Showing 1–50 of 101 results for author: Banerjee, I

Multi-messenger Approach to Ultra-light Scalars

Authors: Indra Kumar Banerjee, Soumya Bonthu, Ujjal Kumar Dey

Abstract: We propose a novel method to study the ultra-light scalars, where compact rotating objects undergo the phenomenon of superradiance to create gravitational waves and neutrino flux signals. The neutrino flux results from the 'right' coupling between the ultra-light scalars and the neutrinos. We study the intertwining of gravitational waves and neutrino flux signals produced from a single source and… ▽ More We propose a novel method to study the ultra-light scalars, where compact rotating objects undergo the phenomenon of superradiance to create gravitational waves and neutrino flux signals. The neutrino flux results from the 'right' coupling between the ultra-light scalars and the neutrinos. We study the intertwining of gravitational waves and neutrino flux signals produced from a single source and elaborate if and when the signals can be detected in existing and upcoming experiments in a direct manner. We also discuss an indirect way to test it by means of cosmic neutrino background which can be detected by upcoming PTOLEMY experiment. △ Less

Submitted 19 June, 2024; originally announced June 2024.

Comments: 16 pages, 5 figures, 1 table

Primordial Black Holes and Gravitational Waves in the $U(1)_{B-L}$ Extended Inert Doublet Model: A First-Order Phase Transition Perspective

Authors: Indra Kumar Banerjee, Ujjal Kumar Dey, Shaaban Khalil

Abstract: We conduct an analysis of a $U(1)_{B-L}$ extended inert doublet model and obtained the parameter space allowing strong first order phase transitions. We show that a large part of the parameter space can cause double first-order phase transitions. Whereas both of these phase transitions can generate a detectable stochastic gravitational wave background, one of them can create primordial black holes… ▽ More We conduct an analysis of a $U(1)_{B-L}$ extended inert doublet model and obtained the parameter space allowing strong first order phase transitions. We show that a large part of the parameter space can cause double first-order phase transitions. Whereas both of these phase transitions can generate a detectable stochastic gravitational wave background, one of them can create primordial black holes with appreciable abundance. The primordial black holes generated at the high scale transition can account for the dark matter maintaining the correct relic abundance. We also show specific benchmark cases and their consequences from the aspect of primordial black holes and gravitational waves. △ Less

Submitted 18 June, 2024; originally announced June 2024.

Comments: 22 pages, 10 figures, 3 tables

Knowledge-grounded Adaptation Strategy for Vision-language Models: Building Unique Case-set for Screening Mammograms for Residents Training

Authors: Aisha Urooj Khan, John Garrett, Tyler Bradshaw, Lonie Salkowski, Jiwoong Jason Jeong, Amara Tariq, Imon Banerjee

Abstract: A visual-language model (VLM) pre-trained on natural images and text pairs poses a significant barrier when applied to medical contexts due to domain shift. Yet, adapting or fine-tuning these VLMs for medical use presents considerable hurdles, including domain misalignment, limited access to extensive datasets, and high-class imbalances. Hence, there is a pressing need for strategies to effectivel… ▽ More A visual-language model (VLM) pre-trained on natural images and text pairs poses a significant barrier when applied to medical contexts due to domain shift. Yet, adapting or fine-tuning these VLMs for medical use presents considerable hurdles, including domain misalignment, limited access to extensive datasets, and high-class imbalances. Hence, there is a pressing need for strategies to effectively adapt these VLMs to the medical domain, as such adaptations would prove immensely valuable in healthcare applications. In this study, we propose a framework designed to adeptly tailor VLMs to the medical domain, employing selective sampling and hard-negative mining techniques for enhanced performance in retrieval tasks. We validate the efficacy of our proposed approach by implementing it across two distinct VLMs: the in-domain VLM (MedCLIP) and out-of-domain VLMs (ALBEF). We assess the performance of these models both in their original off-the-shelf state and after undergoing our proposed training strategies, using two extensive datasets containing mammograms and their corresponding reports. Our evaluation spans zero-shot, few-shot, and supervised scenarios. Through our approach, we observe a notable enhancement in Recall@K performance for the image-text retrieval task. △ Less

Submitted 30 May, 2024; originally announced May 2024.

Assessing Empathy in Large Language Models with Real-World Physician-Patient Interactions

Authors: Man Luo, Christopher J. Warren, Lu Cheng, Haidar M. Abdul-Muhsin, Imon Banerjee

Abstract: The integration of Large Language Models (LLMs) into the healthcare domain has the potential to significantly enhance patient care and support through the development of empathetic, patient-facing chatbots. This study investigates an intriguing question Can ChatGPT respond with a greater degree of empathy than those typically offered by physicians? To answer this question, we collect a de-identifi… ▽ More The integration of Large Language Models (LLMs) into the healthcare domain has the potential to significantly enhance patient care and support through the development of empathetic, patient-facing chatbots. This study investigates an intriguing question Can ChatGPT respond with a greater degree of empathy than those typically offered by physicians? To answer this question, we collect a de-identified dataset of patient messages and physician responses from Mayo Clinic and generate alternative replies using ChatGPT. Our analyses incorporate novel empathy ranking evaluation (EMRank) involving both automated metrics and human assessments to gauge the empathy level of responses. Our findings indicate that LLM-powered chatbots have the potential to surpass human physicians in delivering empathetic communication, suggesting a promising avenue for enhancing patient care and reducing professional burnout. The study not only highlights the importance of empathy in patient interactions but also proposes a set of effective automatic empathy ranking metrics, paving the way for the broader adoption of LLMs in healthcare. △ Less

Submitted 25 May, 2024; originally announced May 2024.

A $π_1$ obstruction to having finite index monodromy and an unusual subgroup of infinite index in $\textrm{Mod}(Σ_g)$

Authors: Ishan Banerjee

Abstract: Let $X$ be an algebraic surface with $\mathcal{L}$ an ample line bundle on $X$. Let $Γ(X, \mathcal{L})$ be the \emph{geometric monodromy} group associated to family of nonsingular curves in $X$ that are zero loci of sections of $\mathcal{L}$. We provide obstructions to $Γ(X, \mathcal{L})$ being finite index in the map** class group. We also show that for any $k \ge 0$, the image of monodromy i… ▽ More Let $X$ be an algebraic surface with $\mathcal{L}$ an ample line bundle on $X$. Let $Γ(X, \mathcal{L})$ be the \emph{geometric monodromy} group associated to family of nonsingular curves in $X$ that are zero loci of sections of $\mathcal{L}$. We provide obstructions to $Γ(X, \mathcal{L})$ being finite index in the map** class group. We also show that for any $k \ge 0$, the image of monodromy is finite index in appropriate subgroups of the quotient of the map** class group by the $k$th term of the Johnson filtration assuming that $\mathcal{L}$ is sufficiently ample. This enables us to construct several subgroups of the map** class group with unusual properties, in some cases providing the first examples of subgroups with those properties. △ Less

Submitted 11 March, 2024; originally announced March 2024.

Error terms for the motives of discriminant complements and a Cayley-Bacharach theorem

Authors: Ishan Banerjee

Abstract: In this paper we prove under some simplifying hypotheses questions of Picoco and Levinson-Ullery on Cayley-Bacharach sets. Our results imply that, under suitable hypotheses Cayley-Bacharach sets lie on curves of low degree. We then use these results to estimate error terms to the normalized motive of the space of smooth degree $d$ hypersurfaces in $\mathbb{P}^n$as $d$ grows to infinity. The error… ▽ More In this paper we prove under some simplifying hypotheses questions of Picoco and Levinson-Ullery on Cayley-Bacharach sets. Our results imply that, under suitable hypotheses Cayley-Bacharach sets lie on curves of low degree. We then use these results to estimate error terms to the normalized motive of the space of smooth degree $d$ hypersurfaces in $\mathbb{P}^n$as $d$ grows to infinity. The error term can be expressed in terms of a certain `sum over points' on plane cubic curves and the associated Hodge structure can be expressed in terms of the cohomology of the moduli space of elliptic curves. We also prove convergence of the motive of degree $d$ hypersurfaces in $\mathbb{P}^n$ as $n$ grows to infinity as well as other results on discriminant complements of high dimensional varieties. △ Less

Submitted 11 March, 2024; originally announced March 2024.

Hierarchical Classification System for Breast Cancer Specimen Report (HCSBC) -- an end-to-end model for characterizing severity and diagnosis

Authors: Thiago Santos, Harish Kamath, Christopher R. McAdams, Mary S. Newell, Marina Mosunjac, Gabriela Oprea-Ilies, Geoffrey Smith, Constance Lehman, Judy Gichoya, Imon Banerjee, Hari Trivedi

Abstract: Automated classification of cancer pathology reports can extract information from unstructured reports and categorize each report into structured diagnosis and severity categories. Thus, such system can reduce the burden for populating tumor registries, help registration for clinical trial as well as develo** large dataset for deep learning model development using true pathologic ground truth. H… ▽ More Automated classification of cancer pathology reports can extract information from unstructured reports and categorize each report into structured diagnosis and severity categories. Thus, such system can reduce the burden for populating tumor registries, help registration for clinical trial as well as develo** large dataset for deep learning model development using true pathologic ground truth. However, the content of breast pathology reports can be difficult for categorize due to the high linguistic variability in content and wide variety of potential diagnoses >50. Existing NLP models are primarily focused on develo** classifier for primary breast cancer types (e.g. IDC, DCIS, ILC) and tumor characteristics, and ignore the rare diagnosis of cancer subtypes. We then developed a hierarchical hybrid transformer-based pipeline (59 labels) - Hierarchical Classification System for Breast Cancer Specimen Report (HCSBC), which utilizes the potential of the transformer context-preserving NLP technique and compared our model to several state of the art ML and DL models. We trained the model on the EUH data and evaluated our model's performance on two external datasets - MGH and Mayo Clinic. We publicly release the code and a live application under Huggingface spaces repository △ Less

Submitted 2 November, 2023; originally announced December 2023.

Spinning Primordial Black Holes from First Order Phase Transition

Authors: Indra Kumar Banerjee, Ujjal Kumar Dey

Abstract: We conduct a novel study to obtain the initial spin of the primordial black holes created during a first-order phase transition due to delayed false vacuum decay. Remaining within the parameter space consistent with observational bounds, we express the abundance and the initial spin of the primordial black holes as functions of the phase transition parameters. The abundance of the primordial black… ▽ More We conduct a novel study to obtain the initial spin of the primordial black holes created during a first-order phase transition due to delayed false vacuum decay. Remaining within the parameter space consistent with observational bounds, we express the abundance and the initial spin of the primordial black holes as functions of the phase transition parameters. The abundance of the primordial black holes is extremely sensitive to the phase transition parameters. We also find that the initial spin weakly depends on all parameters except the transition temperature. △ Less

Submitted 5 November, 2023; originally announced November 2023.

Comments: 9 pages, 5 figures, 1 table

Gravitational Wave Probe of Primordial Black Hole Origin via Superradiance

Authors: Indra Kumar Banerjee, Ujjal Kumar Dey

Abstract: In this article we have used stochastic gravitational wave background as a unique probe to gain insight regarding the creation mechanism of primordial black holes. We have considered the cumulative gravitational wave background which consists of the primary part coming from the creation mechanism of the primordial black holes and the secondary part coming from the different mechanisms the primordi… ▽ More In this article we have used stochastic gravitational wave background as a unique probe to gain insight regarding the creation mechanism of primordial black holes. We have considered the cumulative gravitational wave background which consists of the primary part coming from the creation mechanism of the primordial black holes and the secondary part coming from the different mechanisms the primordial black holes go through. We have shown that in the presence of light or ultra light scalar bosons, superradiant instability generates the secondary part of the gravitational wave background which is the most detectable. In order to show the unique features of the cumulative background, we have considered the delayed vacuum decay during a first order phase transition as the origin of primordial black holes. We have shown the dependence of the features of the cumulative background, such as the mass of the relevant light scalars, peak frequencies, etc. on the transition parameters. We have also generated the cumulative background for a few benchmark cases to further illustrate our claim. △ Less

Submitted 17 April, 2024; v1 submitted 6 November, 2023; originally announced November 2023.

Comments: 18 pages, 8 figures, 1 table; slight change in the title; minor updates in the discussion; references added; matches published version

Journal ref: JCAP 04 (2024) 049

Imprints of Einstein-Maxwell dilaton-axion gravity in the observed shadows of Sgr A* and M87*

Authors: Siddharth Kumar Sahoo, Neeraj Yadav, Indrani Banerjee

Abstract: Einstein-Maxwell dilaton-axion (EMDA) gravity provides a simple framework to investigate the signatures of string theory. The axion and the dilaton fields arising in EMDA gravity have important implications in inflationary cosmology and in addressing the late time acceleration of the universe. It is therefore instructive to explore the implications of such a model in explaining the astrophysical o… ▽ More Einstein-Maxwell dilaton-axion (EMDA) gravity provides a simple framework to investigate the signatures of string theory. The axion and the dilaton fields arising in EMDA gravity have important implications in inflationary cosmology and in addressing the late time acceleration of the universe. It is therefore instructive to explore the implications of such a model in explaining the astrophysical observations. In this work we explore the role of EMDA gravity in explaining the observed shadows of black holes (M87* and Sgr A*) released by the Event Horizon Telescope (EHT) collaboration. The Kerr-Sen metric represents the exact, stationary and axisymmetric black hole solution of EMDA gravity. Such a black hole is characterized by the angular momentum $a$ acquired from the axionic field and the dilatonic charge $r_2$ arising from string compactifications. We study the role of spin and the dilaton charge in modifying the shape and size of the black hole shadow. We note that black holes with larger dilaton charge cast a smaller shadow. We investigate the consequences of such a result in addressing the EHT observations of M87* and Sgr A*. Our analysis reveals that the shadow of M87* exhibits a preference towards the Kerr scenario. However, when 10% offset in the shadow diameter is considered, $0.1\lesssim r_2\lesssim 0.3$ is observationally favored within 1-$σ$. The shadow of Sgr A* on the other hand shows a preference towards the Kerr-Sen scenario since the central value of its shadow can be better explained by a non-zero dilaton charge $0.1 \lesssim r_2 \lesssim 0.4$. However, when the 1-$σ$ interval is considered the Kerr scenario is included. We discuss the implications of our results. △ Less

Submitted 24 May, 2023; originally announced May 2023.

Comments: 28 pages, 7 figures

Probing the Origin of Primordial Black Holes through Novel Gravitational Wave Spectrum

Authors: Indra Kumar Banerjee, Ujjal Kumar Dey

Abstract: In this article we investigate the cumulative stochastic gravitational wave spectra as a tool to gain insight on the creation mechanism of primordial black holes. We consider gravitational waves from the production mechanism of primordial black holes and from the gravitational interactions of those primordial black holes among themselves and other astrophysical black holes. We specifically focus o… ▽ More In this article we investigate the cumulative stochastic gravitational wave spectra as a tool to gain insight on the creation mechanism of primordial black holes. We consider gravitational waves from the production mechanism of primordial black holes and from the gravitational interactions of those primordial black holes among themselves and other astrophysical black holes. We specifically focus on asynchronous bubble nucleation during a first order phase transition as the creation mechanism. We have used two benchmark phase transitions through which the primordial black holes and the primary gravitational wave spectra have been generated. We have considered binary systems and close hyperbolic interactions of primordial black holes with other primordial and astrophysical black holes as the source of the secondary part of the spectra. We have shown that this unique cumulative spectra have features which directly and indirectly depend on the specifics of the production mechanism. △ Less

Submitted 13 July, 2023; v1 submitted 12 May, 2023; originally announced May 2023.

Comments: 22 pages, 9 figures, 2 tables; figures updated, discussions and references added, version published in JCAP

Journal ref: JCAP 07 (2023) 024

Multivariate Analysis on Performance Gaps of Artificial Intelligence Models in Screening Mammography

Authors: Linglin Zhang, Beatrice Brown-Mulry, Vineela Nalla, InChan Hwang, Judy Wawira Gichoya, Aimilia Gastounioti, Imon Banerjee, Laleh Seyyed-Kalantari, MinJae Woo, Hari Trivedi

Abstract: Although deep learning models for abnormality classification can perform well in screening mammography, the demographic, imaging, and clinical characteristics associated with increased risk of model failure remain unclear. This retrospective study uses the Emory BrEast Imaging Dataset(EMBED) containing mammograms from 115931 patients imaged at Emory Healthcare between 2013-2020, with BI-RADS asses… ▽ More Although deep learning models for abnormality classification can perform well in screening mammography, the demographic, imaging, and clinical characteristics associated with increased risk of model failure remain unclear. This retrospective study uses the Emory BrEast Imaging Dataset(EMBED) containing mammograms from 115931 patients imaged at Emory Healthcare between 2013-2020, with BI-RADS assessment, region of interest coordinates for abnormalities, imaging features, pathologic outcomes, and patient demographics. Multiple deep learning models were trained to distinguish between abnormal tissue patches and randomly selected normal tissue patches from screening mammograms. We assessed model performance by subgroups defined by age, race, pathologic outcome, tissue density, and imaging characteristics and investigated their associations with false negatives (FN) and false positives (FP). We also performed multivariate logistic regression to control for confounding between subgroups. The top-performing model, ResNet152V2, achieved accuracy of 92.6%(95%CI=92.0-93.2%), and AUC 0.975(95%CI=0.972-0.978). Before controlling for confounding, nearly all subgroups showed statistically significant differences in model performance. However, after controlling for confounding, we found lower FN risk associates with Other race(RR=0.828;p=.050), biopsy-proven benign lesions(RR=0.927;p=.011), and mass(RR=0.921;p=.010) or asymmetry(RR=0.854;p=.040); higher FN risk associates with architectural distortion (RR=1.037;p<.001). Higher FP risk associates to BI-RADS density C(RR=1.891;p<.001) and D(RR=2.486;p<.001). Our results demonstrate subgroup analysis is important in mammogram classifier performance evaluation, and controlling for confounding between subgroups elucidates the true associations between variables and model failure. These results can help guide develo** future breast cancer detection models. △ Less

Submitted 19 October, 2023; v1 submitted 7 May, 2023; originally announced May 2023.

Comments: 29 pages, 6 tables, 7 figures, 2 supplemental tables

PTOLEMY's test of generalized neutrino interactions: unveiling challenges and constraints

Authors: Indra Kumar Banerjee, Ujjal Kumar Dey, Newton Nath, Saadat Salman Shariff

Abstract: Unanswered questions surrounding neutrinos have motivated investigations into physics beyond the standard model (SM) of particle physics. In particular, generalized neutrino interactions (GNI) provide a broader framework for studying these effects compared to the commonly studied non-standard neutrino interactions. These interactions are described by higher dimensional operators while maintaining… ▽ More Unanswered questions surrounding neutrinos have motivated investigations into physics beyond the standard model (SM) of particle physics. In particular, generalized neutrino interactions (GNI) provide a broader framework for studying these effects compared to the commonly studied non-standard neutrino interactions. These interactions are described by higher dimensional operators while maintaining the gauge symmetries of the SM. Furthermore, the cosmic neutrino background, a predicted component of the SM and standard cosmology, has yet to be directly detected. To shed light on this elusive phenomenon, we conduct a comprehensive analysis of the relevant GNI, specifically focusing on their implications for the proposed cosmic neutrino detector PTOLEMY. We make an attempt to see the capabilities and the limitations of PTOLEMY in sensing GNI while remaining optimistic regarding PTOLEMY's experimental resolution. These interactions play a significant role in modifying the electron spectrum resulting from the capture of cosmic neutrinos on radioactive tritium. This work also explores how the presence of these interactions influences the differential electron spectrum, taking into account factors such as finite experimental resolution, the mass of the lightest neutrino eigenstate, the strength of the interactions, and the ordering of neutrino mass. △ Less

Submitted 3 April, 2024; v1 submitted 5 April, 2023; originally announced April 2023.

Comments: 23 pages, 5 figures, 5 tables; slight change in the title; major updates in the discussion; results essentially unchanged; references added; matches published version

Journal ref: JCAP 04 (2024) 002

MLOps with enhanced performance control and observability

Authors: Indradumna Banerjee, Dinesh Ghanta, Girish Nautiyal, Pradeep Sanchana, Prateek Katageri, Atin Modi

Abstract: The explosion of data and its ever increasing complexity in the last few years, has made MLOps systems more prone to failure, and new tools need to be embedded in such systems to avoid such failure. In this demo, we will introduce crucial tools in the observability module of a MLOps system that target difficult issues like data drfit and model version control for optimum model selection. We believ… ▽ More The explosion of data and its ever increasing complexity in the last few years, has made MLOps systems more prone to failure, and new tools need to be embedded in such systems to avoid such failure. In this demo, we will introduce crucial tools in the observability module of a MLOps system that target difficult issues like data drfit and model version control for optimum model selection. We believe integrating these features in our MLOps pipeline would go a long way in building a robust system immune to early stage ML system failures. △ Less

Submitted 2 February, 2023; originally announced February 2023.

Comments: SECOND INTERNATIONAL CONFERENCE ON AI-ML SYSTEMS

Ngram-LSTM Open Rate Prediction Model (NLORP) and Error_accuracy@C metric: Simple effective, and easy to implement approach to predict open rates for marketing email

Authors: Shubham Joshi, Indradumna Banerjee

Abstract: Our generation has seen an exponential increase in digital tools adoption. One of the unique areas where digital tools have made an exponential foray is in the sphere of digital marketing, where goods and services have been extensively promoted through the use of digital advertisements. Following this growth, multiple companies have leveraged multiple apps and channels to display their brand ident… ▽ More Our generation has seen an exponential increase in digital tools adoption. One of the unique areas where digital tools have made an exponential foray is in the sphere of digital marketing, where goods and services have been extensively promoted through the use of digital advertisements. Following this growth, multiple companies have leveraged multiple apps and channels to display their brand identities to a significantly larger user base. This has resulted in products, worth billions of dollars to be sold online. Emails and push notifications have become critical channels to publish advertisement content, to proactively engage with their contacts. Several marketing tools provide a user interface for marketers to design Email and Push messages for digital marketing campaigns. Marketers are also given a predicted open rate for the entered subject line. For enabling marketers generate targeted subject lines, multiple machine learning techniques have been used in the recent past. In particular, deep learning techniques that have established good effectiveness and efficiency. However, these techniques require a sizable amount of labelled training data in order to get good results. The creation of such datasets, particularly those with subject lines that have a specific theme, is a challenging and time-consuming task. In this paper, we propose a novel Ngram and LSTM-based modeling approach (NLORPM) to predict open rates of entered subject lines that is easier to implement, has low prediction latency, and performs extremely well for sparse data. To assess the performance of this model, we also devise a new metric called 'Error_accuracy@C' which is simple to grasp and fully comprehensible to marketers. △ Less

Submitted 14 February, 2023; v1 submitted 25 January, 2023; originally announced February 2023.

Generalizable Natural Language Processing Framework for Migraine Reporting from Social Media

Authors: Yuting Guo, Swati Rajwal, Sahithi Lakamana, Chia-Chun Chiang, Paul C. Menell, Adnan H. Shahid, Yi-Chieh Chen, Nikita Chhabra, Wan-Ju Chao, Chieh-Ju Chao, Todd J. Schwedt, Imon Banerjee, Abeed Sarker

Abstract: Migraine is a high-prevalence and disabling neurological disorder. However, information migraine management in real-world settings could be limited to traditional health information sources. In this paper, we (i) verify that there is substantial migraine-related chatter available on social media (Twitter and Reddit), self-reported by migraine sufferers; (ii) develop a platform-independent text cla… ▽ More Migraine is a high-prevalence and disabling neurological disorder. However, information migraine management in real-world settings could be limited to traditional health information sources. In this paper, we (i) verify that there is substantial migraine-related chatter available on social media (Twitter and Reddit), self-reported by migraine sufferers; (ii) develop a platform-independent text classification system for automatically detecting self-reported migraine-related posts, and (iii) conduct analyses of the self-reported posts to assess the utility of social media for studying this problem. We manually annotated 5750 Twitter posts and 302 Reddit posts. Our system achieved an F1 score of 0.90 on Twitter and 0.93 on Reddit. Analysis of information posted by our 'migraine cohort' revealed the presence of a plethora of relevant information about migraine therapies and patient sentiments associated with them. Our study forms the foundation for conducting an in-depth analysis of migraine-related information using social media data. △ Less

Submitted 23 December, 2022; originally announced December 2022.

Comments: Accepted by AMIA 2023 Informatics Summit

Sums of Two Squares Visualized

Authors: Ishan Banerjee, Amites Sarkar

Abstract: We provide a geometric interpretation of Brillhart's celebrated algorithm for expressing a prime $p\equiv 1\pmod 4$ as the sum of two squares. We provide a geometric interpretation of Brillhart's celebrated algorithm for expressing a prime $p\equiv 1\pmod 4$ as the sum of two squares. △ Less

Submitted 13 December, 2022; originally announced December 2022.

Comments: 8 pages, 3 figures

MSC Class: 01A45; 11A99

Offline Estimation of Controlled Markov Chains: Minimaxity and Sample Complexity

Authors: Imon Banerjee, Harsha Honnappa, Vinayak Rao

Abstract: In this work, we study a natural nonparametric estimator of the transition probability matrices of a finite controlled Markov chain. We consider an offline setting with a fixed dataset, collected using a so-called logging policy. We develop sample complexity bounds for the estimator and establish conditions for minimaxity. Our statistical bounds depend on the logging policy through its mixing prop… ▽ More In this work, we study a natural nonparametric estimator of the transition probability matrices of a finite controlled Markov chain. We consider an offline setting with a fixed dataset, collected using a so-called logging policy. We develop sample complexity bounds for the estimator and establish conditions for minimaxity. Our statistical bounds depend on the logging policy through its mixing properties. We show that achieving a particular statistical risk bound involves a subtle and interesting trade-off between the strength of the mixing properties and the number of samples. We demonstrate the validity of our results under various examples, such as ergodic Markov chains, weakly ergodic inhomogeneous Markov chains, and controlled Markov chains with non-stationary Markov, episodic, and greedy controls. Lastly, we use these sample complexity bounds to establish concomitant ones for offline evaluation of stationary Markov control policies. △ Less

Submitted 26 January, 2024; v1 submitted 13 November, 2022; originally announced November 2022.

Comments: 71 pages, 23 main

Neutrino Decoherence from Generalised Uncertainty

Authors: Indra Kumar Banerjee, Ujjal Kumar Dey

Abstract: Quantum gravity models predict a minimal measurable length which gives rise to a modification in the uncertainty principle. One of the simplest manifestations of these generalised uncertainty principles is the linear quadratic generalised uncertainty principle which leads to a modified Heisenberg algebra. This can alter the usual von-Neumann evolution of density matrix to a Lindblad-type equation.… ▽ More Quantum gravity models predict a minimal measurable length which gives rise to a modification in the uncertainty principle. One of the simplest manifestations of these generalised uncertainty principles is the linear quadratic generalised uncertainty principle which leads to a modified Heisenberg algebra. This can alter the usual von-Neumann evolution of density matrix to a Lindblad-type equation. We show how this can give rise to a decoherence in neutrino propagation in vacuum. The decoherence effects due to the linear quadratic generalised uncertainty principle are extremely minimal and is unlikely to be detectable in the existing or upcoming experimental facilities for any of the natural sources of neutrinos. We also show that, in principle, there can be other variants of generalised uncertainty principle which predicts verifiable decoherence effects for the cosmic neutrino background. △ Less

Submitted 12 May, 2023; v1 submitted 25 August, 2022; originally announced August 2022.

Comments: 20 pages, 3 figures; updated formalism; new discussions added; figure and reference list updated; to appear in Eur. Phys. J. C

Journal ref: Eur. Phys. J. C (2023) 83, 428

Meta Sparse Principal Component Analysis

Authors: Imon Banerjee, Jean Honorio

Abstract: We study the meta-learning for support (i.e. the set of non-zero entries) recovery in high-dimensional Principal Component Analysis. We reduce the sufficient sample complexity in a novel task with the information that is learned from auxiliary tasks. We assume each task to be a different random Principal Component (PC) matrix with a possibly different support and that the support union of the PC m… ▽ More We study the meta-learning for support (i.e. the set of non-zero entries) recovery in high-dimensional Principal Component Analysis. We reduce the sufficient sample complexity in a novel task with the information that is learned from auxiliary tasks. We assume each task to be a different random Principal Component (PC) matrix with a possibly different support and that the support union of the PC matrices is small. We then pool the data from all the tasks to execute an improper estimation of a single PC matrix by maximising the $l_1$-regularised predictive covariance to establish that with high probability the true support union can be recovered provided a sufficient number of tasks $m$ and a sufficient number of samples $ O\left(\frac{\log(p)}{m}\right)$ for each task, for $p$-dimensional vectors. Then, for a novel task, we prove that the maximisation of the $l_1$-regularised predictive covariance with the additional constraint that the support is a subset of the estimated support union could reduce the sufficient sample complexity of successful support recovery to $O(\log |J|)$, where $J$ is the support union recovered from the auxiliary tasks. Typically, $|J|$ would be much less than $p$ for sparse matrices. Finally, we demonstrate the validity of our experiments through numerical simulations. △ Less

Submitted 19 August, 2022; v1 submitted 18 August, 2022; originally announced August 2022.

Comments: 29 pages, 7 figures

Rotating hairy black holes and thermodynamics from gravitational decoupling

Authors: Subhash Mahapatra, Indrani Banerjee

Abstract: We study the method of extended gravitational decoupling in obtaining static black hole solutions satisfying Einstein's equations with a tensor vacuum. The source has quite generic characteristics and satisfies the strong energy condition. The stationary, axisymmetric counterpart of the static metric is obtained by applying the Newman-Janis and Azreg-Aïnou algorithms. The thermodynamics of the rot… ▽ More We study the method of extended gravitational decoupling in obtaining static black hole solutions satisfying Einstein's equations with a tensor vacuum. The source has quite generic characteristics and satisfies the strong energy condition. The stationary, axisymmetric counterpart of the static metric is obtained by applying the Newman-Janis and Azreg-Aïnou algorithms. The thermodynamics of the rotating solution is studied and the expressions of various thermodynamic quantities are derived. The dependence of the temperature, free energy and specific heat on the horizon radius is studied for various values of the hairy parameter and the black hole spin. Such a study reveals that small hairy black holes are thermodynamically more stable compared to large hairy black holes, and that the horizon radius and temperature range for which the rotating hairy black holes can be in thermodynamic equilibrium with the surroundings depends non-trivially on the hairy parameters. We further discuss the first law of black hole thermodynamics for the hairy case and discuss its implications. △ Less

Submitted 24 January, 2023; v1 submitted 11 August, 2022; originally announced August 2022.

Comments: 30 Pages, 17 Figures, references added, discussion extended, Published version

Journal ref: Phys. Dark Univ. 39 (2023) 101172

Augmenting Vision Language Pretraining by Learning Codebook with Visual Semantics

Authors: Xiaoyuan Guo, Jiali Duan, C. -C. Jay Kuo, Judy Wawira Gichoya, Imon Banerjee

Abstract: Language modality within the vision language pretraining framework is innately discretized, endowing each word in the language vocabulary a semantic meaning. In contrast, visual modality is inherently continuous and high-dimensional, which potentially prohibits the alignment as well as fusion between vision and language modalities. We therefore propose to "discretize" the visual representation by… ▽ More Language modality within the vision language pretraining framework is innately discretized, endowing each word in the language vocabulary a semantic meaning. In contrast, visual modality is inherently continuous and high-dimensional, which potentially prohibits the alignment as well as fusion between vision and language modalities. We therefore propose to "discretize" the visual representation by joint learning a codebook that imbues each visual token a semantic. We then utilize these discretized visual semantics as self-supervised ground-truths for building our Masked Image Modeling objective, a counterpart of Masked Language Modeling which proves successful for language models. To optimize the codebook, we extend the formulation of VQ-VAE which gives a theoretic guarantee. Experiments validate the effectiveness of our approach across common vision-language benchmarks. △ Less

Submitted 31 July, 2022; originally announced August 2022.

Comments: 7 pages, 4 figures, ICPR2022. arXiv admin note: text overlap with arXiv:2203.00048

Hunting extra dimensions in the shadow of Sgr A*

Authors: Indrani Banerjee, Sumanta Chakraborty, Soumitra SenGupta

Abstract: We show that the observed angular diameter of the shadow of the ultra compact object Sgr A*, favours the existence of an extra spatial dimension. This holds irrespective of the nature of the ultra compact object, i.e., whether it is a wormhole or, a black hole mimicker, but with the common feature that both of them have an extra dimensional origin. This result holds true for the mass and the dista… ▽ More We show that the observed angular diameter of the shadow of the ultra compact object Sgr A*, favours the existence of an extra spatial dimension. This holds irrespective of the nature of the ultra compact object, i.e., whether it is a wormhole or, a black hole mimicker, but with the common feature that both of them have an extra dimensional origin. This result holds true for the mass and the distance measurements of Sgr A* using both Keck and the Gravity collaborations and whether we use the observed image or, the observed shadow diameter. In particular, the central value of the observed shadow or, the observed image diameter predicts non-zero hairs inherited from the extra dimensions. △ Less

Submitted 5 December, 2022; v1 submitted 18 July, 2022; originally announced July 2022.

Comments: v2, published version, 12 pages, 4 figures

Journal ref: Phys. Rev. D 106, 084051 (2022)

Do shadows of Sgr A* and M87* indicate black holes with a magnetic monopole charge?

Authors: Indrani Banerjee, Subhadip Sau, Soumitra SenGupta

Abstract: We study the prospect of Bardeen black holes in explaining the observed shadow of Sgr A* and M87*. Bardeen black holes are regular black holes endowed with a magnetic monopole charge that arise in Einstein gravity coupled to non-linear electrodynamics. These black holes are interesting as they can evade the r = 0 curvature singularity arising in general relativity. It is therefore worthwhile to lo… ▽ More We study the prospect of Bardeen black holes in explaining the observed shadow of Sgr A* and M87*. Bardeen black holes are regular black holes endowed with a magnetic monopole charge that arise in Einstein gravity coupled to non-linear electrodynamics. These black holes are interesting as they can evade the r = 0 curvature singularity arising in general relativity. It is therefore worthwhile to look for signatures of Bardeen black holes in astrophysical observations. With two successive release of black hole images by the Event Horizon Telescope (EHT) collaboration, the scope to test the nature of strong gravity has substantially increased. We compare the theoretically computed shadow observables with the observed image of Sgr A* and M87*. Our analysis reveals that while the observed angular diameter of M87* favors the Kerr scenario, the shadow of Sgr A* can be better explained by the Bardeen background. This indicates that although rare, certain black holes exhibit a preference towards regular black holes like the Bardeen spacetime. △ Less

Submitted 13 July, 2022; originally announced July 2022.

Comments: 24 pages, 8 figures

Fitness Dependent Optimizer for IoT Healthcare using Adapted Parameters: A Case Study Implementation

Authors: Aso M. Aladdin, Jaza M. Abdullah, Kazhan Othman Mohammed Salih, Tarik A. Rashid, Rafid Sagban, Abeer Alsaddon, Nebojsa Bacanin, Amit Chhabra, S. Vimal, Indradip Banerjee

Abstract: This discusses a case study on Fitness Dependent Optimizer or so-called FDO and adapting its parameters to the Internet of Things (IoT) healthcare. The reproductive way is sparked by the bee swarm and the collaborative decision-making of FDO. As opposed to the honey bee or artificial bee colony algorithms, this algorithm has no connection to them. In FDO, the search agent's position is updated usi… ▽ More This discusses a case study on Fitness Dependent Optimizer or so-called FDO and adapting its parameters to the Internet of Things (IoT) healthcare. The reproductive way is sparked by the bee swarm and the collaborative decision-making of FDO. As opposed to the honey bee or artificial bee colony algorithms, this algorithm has no connection to them. In FDO, the search agent's position is updated using speed or velocity, but it's done differently. It creates weights based on the fitness function value of the problem, which assists lead the agents through the exploration and exploitation processes. Other algorithms are evaluated and compared to FDO as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) in the original work. The key current algorithms:The Salp-Swarm Algorithms (SSA), Dragonfly Algorithm (DA), and Whale Optimization Algorithm (WOA) have been evaluated against FDO in terms of their results. Using these FDO experimental findings, we may conclude that FDO outperforms the other techniques stated. There are two primary goals for this chapter: first, the implementation of FDO will be shown step-by-step so that readers can better comprehend the algorithm method and apply FDO to solve real-world applications quickly. The second issue deals with how to tweak the FDO settings to make the meta-heuristic evolutionary algorithm better in the IoT health service system at evaluating big quantities of information. Ultimately, the target of this chapter's enhancement is to adapt the IoT healthcare framework based on FDO to spawn effective IoT healthcare applications for reasoning out real-world optimization, aggregation, prediction, segmentation, and other technological problems. △ Less

Submitted 18 May, 2022; originally announced July 2022.

Comments: 17 pages

Journal ref: -

Advances in Prediction of Readmission Rates Using Long Term Short Term Memory Networks on Healthcare Insurance Data

Authors: Shuja Khalid, Francisco Matos, Ayman Abunimer, Joel Bartlett, Richard Duszak, Michal Horny, Judy Gichoya, Imon Banerjee, Hari Trivedi

Abstract: 30-day hospital readmission is a long standing medical problem that affects patients' morbidity and mortality and costs billions of dollars annually. Recently, machine learning models have been created to predict risk of inpatient readmission for patients with specific diseases, however no model exists to predict this risk across all patients. We developed a bi-directional Long Short Term Memory (… ▽ More 30-day hospital readmission is a long standing medical problem that affects patients' morbidity and mortality and costs billions of dollars annually. Recently, machine learning models have been created to predict risk of inpatient readmission for patients with specific diseases, however no model exists to predict this risk across all patients. We developed a bi-directional Long Short Term Memory (LSTM) Network that is able to use readily available insurance data (inpatient visits, outpatient visits, and drug prescriptions) to predict 30 day re-admission for any admitted patient, regardless of reason. The top-performing model achieved an ROC AUC of 0.763 (0.011) when using historical, inpatient, and post-discharge data. The LSTM model significantly outperformed a baseline random forest classifier, indicating that understanding the sequence of events is important for model prediction. Incorporation of 30-days of historical data also significantly improved model performance compared to inpatient data alone, indicating that a patients clinical history prior to admission, including outpatient visits and pharmacy data is a strong contributor to readmission. Our results demonstrate that a machine learning model is able to predict risk of inpatient readmission with reasonable accuracy for all patients using structured insurance billing data. Because billing data or equivalent surrogates can be extracted from sites, such a model could be deployed to identify patients at risk for readmission before they are discharged, or to assign more robust follow up (closer follow up, home health, mailed medications) to at-risk patients after discharge. △ Less

Submitted 30 June, 2022; originally announced July 2022.

Comments: 7 pages, 3 figures, 3 tables

Signatures of regular black holes from the shadow of Sgr A* and M87*

Authors: Indrani Banerjee, Subhadip Sau, Soumitra SenGupta

Abstract: With the recent release of the black hole image of Sgr A* alongside the earlier image of M87*, one can now really hope to acquire a better understanding of the gravitational physics at the horizon scale. In this paper, we investigate the prospect of the regular black hole scenario with a Minkowski core in explaining the observed shadow of M87* and Sgr A*. Regular black holes generally appear in Ei… ▽ More With the recent release of the black hole image of Sgr A* alongside the earlier image of M87*, one can now really hope to acquire a better understanding of the gravitational physics at the horizon scale. In this paper, we investigate the prospect of the regular black hole scenario with a Minkowski core in explaining the observed shadow of M87* and Sgr A*. Regular black holes generally appear in Einstein gravity coupled to non-linear electrodynamics and are interesting as they can evade the r = 0 curvature singularity arising in general relativity. Using the previously determined mass and distance we compute the observables associated with the black hole shadow. These when compared with the observed angular diameter reveal that the shadow of M87* and Sgr A* favor the regular black hole scenario with a small but non-zero charge. The implications are discussed. △ Less

Submitted 24 September, 2022; v1 submitted 24 June, 2022; originally announced June 2022.

Comments: 23 pages, 9 figures, matches published version (JCAP accepted)

Signatures of regular black holes from the quasar continuum spectrum

Authors: Indrani Banerjee

Abstract: Regular black holes arising in Einstein gravity coupled to non-linear electrodynamics are worth studying as they can circumvent the r = 0 curvature singularity arising in general relativity. In this work we explore the signatures of regular black holes with a Minkowski core from the quasar continuum spectrum. We use thin-disk approximation to derive the theoretical luminosity from the accretion di… ▽ More Regular black holes arising in Einstein gravity coupled to non-linear electrodynamics are worth studying as they can circumvent the r = 0 curvature singularity arising in general relativity. In this work we explore the signatures of regular black holes with a Minkowski core from the quasar continuum spectrum. We use thin-disk approximation to derive the theoretical luminosity from the accretion disk and compare it with the optical data of eighty Palomar Green quasars. Our analysis based on error estimators like the chi-square, the Nash-Sutcliffe efficiency, the index of agreement etc. reveal that optical observations of quasars favor the Kerr scenario compared to black holes in non-linear electrodynamics. The implications are discussed. △ Less

Submitted 14 June, 2022; originally announced June 2022.

Comments: 22 pages, 6 figures

PathologyBERT -- Pre-trained Vs. A New Transformer Language Model for Pathology Domain

Authors: Thiago Santos, Amara Tariq, Susmita Das, Kavyasree Vayalpati, Geoffrey H. Smith, Hari Trivedi, Imon Banerjee

Abstract: Pathology text mining is a challenging task given the reporting variability and constant new findings in cancer sub-type definitions. However, successful text mining of a large pathology database can play a critical role to advance 'big data' cancer research like similarity-based treatment selection, case identification, prognostication, surveillance, clinical trial screening, risk stratification,… ▽ More Pathology text mining is a challenging task given the reporting variability and constant new findings in cancer sub-type definitions. However, successful text mining of a large pathology database can play a critical role to advance 'big data' cancer research like similarity-based treatment selection, case identification, prognostication, surveillance, clinical trial screening, risk stratification, and many others. While there is a growing interest in develo** language models for more specific clinical domains, no pathology-specific language space exist to support the rapid data-mining development in pathology space. In literature, a few approaches fine-tuned general transformer models on specialized corpora while maintaining the original tokenizer, but in fields requiring specialized terminology, these models often fail to perform adequately. We propose PathologyBERT - a pre-trained masked language model which was trained on 347,173 histopathology specimen reports and publicly released in the Huggingface repository. Our comprehensive experiments demonstrate that pre-training of transformer model on pathology corpora yields performance improvements on Natural Language Understanding (NLU) and Breast Cancer Diagnose Classification when compared to nonspecific language models. △ Less

Submitted 13 May, 2022; originally announced May 2022.

Comments: submitted to "American Medical Informatics Association (AMIA)" 2022 Annual Symposium

Aspects of non-singular bounce in modified gravity theories

Authors: Indrani Banerjee, Tanmoy Paul, Soumitra SenGupta

Abstract: Scenario of a bouncing universe is one of the most active area of research to arrive at singularity free cosmological models. Different proposals have been suggested to avoid the so called 'big bang' singularity through the quantum aspect of gravity which is yet to have a proper understanding. In this work, on the contrary, we consider three different approaches, each of which goes beyond General… ▽ More Scenario of a bouncing universe is one of the most active area of research to arrive at singularity free cosmological models. Different proposals have been suggested to avoid the so called 'big bang' singularity through the quantum aspect of gravity which is yet to have a proper understanding. In this work, on the contrary, we consider three different approaches, each of which goes beyond General Relativity but remain within the domain of classical cosmological scenario, to address this problem. The hallmark of all these approaches is that the origin of the bouncing mechanism is somewhat natural within the geometrical framework of the model without any need of incorporating external source by hand. In the context of these scenarios ,we also discuss various constraints that these viable cosmological models need to satisfy. △ Less

Submitted 28 August, 2022; v1 submitted 11 May, 2022; originally announced May 2022.

Comments: Accepted by GRG in the memorial volume for Prof. Thanu Padmanabhan. arXiv admin note: text overlap with arXiv:2109.00345, arXiv:2202.13186

Multimodal spatiotemporal graph neural networks for improved prediction of 30-day all-cause hospital readmission

Authors: Siyi Tang, Amara Tariq, Jared Dunnmon, Umesh Sharma, Praneetha Elugunti, Daniel Rubin, Bhavik N. Patel, Imon Banerjee

Abstract: Measures to predict 30-day readmission are considered an important quality factor for hospitals as accurate predictions can reduce the overall cost of care by identifying high risk patients before they are discharged. While recent deep learning-based studies have shown promising empirical results on readmission prediction, several limitations exist that may hinder widespread clinical utility, such… ▽ More Measures to predict 30-day readmission are considered an important quality factor for hospitals as accurate predictions can reduce the overall cost of care by identifying high risk patients before they are discharged. While recent deep learning-based studies have shown promising empirical results on readmission prediction, several limitations exist that may hinder widespread clinical utility, such as (a) only patients with certain conditions are considered, (b) existing approaches do not leverage data temporality, (c) individual admissions are assumed independent of each other, which is unrealistic, (d) prior studies are usually limited to single source of data and single center data. To address these limitations, we propose a multimodal, modality-agnostic spatiotemporal graph neural network (MM-STGNN) for prediction of 30-day all-cause hospital readmission that fuses multimodal in-patient longitudinal data. By training and evaluating our methods using longitudinal chest radiographs and electronic health records from two independent centers, we demonstrate that MM-STGNN achieves AUROC of 0.79 on both primary and external datasets. Furthermore, MM-STGNN significantly outperforms the current clinical reference standard, LACE+ score (AUROC=0.61), on the primary dataset. For subset populations of patients with heart and vascular disease, our model also outperforms baselines on predicting 30-day readmission (e.g., 3.7 point improvement in AUROC in patients with heart disease). Lastly, qualitative model interpretability analysis indicates that while patients' primary diagnoses were not explicitly used to train the model, node features crucial for model prediction directly reflect patients' primary diagnoses. Importantly, our MM-STGNN is agnostic to node feature modalities and could be utilized to integrate multimodal data for triaging patients in various downstream resource allocation tasks. △ Less

Submitted 14 April, 2022; originally announced April 2022.

Journal ref: IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 4, pp. 2071-2082, April 2023

OSCARS: An Outlier-Sensitive Content-Based Radiography Retrieval System

Authors: Xiaoyuan Guo, Jiali Duan, Saptarshi Purkayastha, Hari Trivedi, Judy Wawira Gichoya, Imon Banerjee

Abstract: Improving the retrieval relevance on noisy datasets is an emerging need for the curation of a large-scale clean dataset in the medical domain. While existing methods can be applied for class-wise retrieval (aka. inter-class), they cannot distinguish the granularity of likeness within the same class (aka. intra-class). The problem is exacerbated on medical external datasets, where noisy samples of… ▽ More Improving the retrieval relevance on noisy datasets is an emerging need for the curation of a large-scale clean dataset in the medical domain. While existing methods can be applied for class-wise retrieval (aka. inter-class), they cannot distinguish the granularity of likeness within the same class (aka. intra-class). The problem is exacerbated on medical external datasets, where noisy samples of the same class are treated equally during training. Our goal is to identify both intra/inter-class similarities for fine-grained retrieval. To achieve this, we propose an Outlier-Sensitive Content-based rAdiologhy Retrieval System (OSCARS), consisting of two steps. First, we train an outlier detector on a clean internal dataset in an unsupervised manner. Then we use the trained detector to generate the anomaly scores on the external dataset, whose distribution will be used to bin intra-class variations. Second, we propose a quadruplet (a, p, nintra, ninter) sampling strategy, where intra-class negatives nintra are sampled from bins of the same class other than the bin anchor a belongs to, while niner are randomly sampled from inter-classes. We suggest a weighted metric learning objective to balance the intra and inter-class feature learning. We experimented on two representative public radiography datasets. Experiments show the effectiveness of our approach. The training and evaluation code can be found in https://github.com/XiaoyuanGuo/oscars. △ Less

Submitted 6 April, 2022; originally announced April 2022.

Comments: 12 pages, 6 figures, 2 tables

Testing black holes in non-linear electrodynamics from the observed quasi-periodic oscillations

Authors: Indrani Banerjee

Abstract: Quasi-periodic oscillations (QPOs), in particular, the ones with high frequencies, often observed in the power spectrum of black holes, are useful in understanding the nature of strong gravity since they are associated with the motion of matter in the vicinity of the black hole horizon. Interestingly, these high frequency QPOs (HFQPOs) are observed in commensurable pairs, the most common ratio bei… ▽ More Quasi-periodic oscillations (QPOs), in particular, the ones with high frequencies, often observed in the power spectrum of black holes, are useful in understanding the nature of strong gravity since they are associated with the motion of matter in the vicinity of the black hole horizon. Interestingly, these high frequency QPOs (HFQPOs) are observed in commensurable pairs, the most common ratio being 3:2. Several theoretical models are proposed in the literature which explain the HFQPOs in terms of the orbital and epicyclic frequencies of matter rotating around the central object. Since these frequencies are sensitive to the background spacetime, the observed HFQPOs can potentially extract useful information regarding the nature of the same. In this work, we investigate the role of regular black holes with a Minkowski core, which arise in gravity coupled to non-linear electrodynamics, in explaining the HFQPOs. Regular black holes are particularly interesting as they provide a possible resolution to the singularity problem in general relativity. We compare the model dependent QPO frequencies with the available observations of the quasi-periodic oscillations from black hole sources and perform a \c{hi} 2 analysis. Our study reveals that most QPO models favor small but non-trivial values of the non-linear electrodynamics charge parameter. In particular, black holes with large values of non-linear electrodynamics charge parameter are generically disfavored by present observations related to QPOs. △ Less

Submitted 24 September, 2022; v1 submitted 21 March, 2022; originally announced March 2022.

Comments: 24 pages, 2 figures, 4 tables. arXiv admin note: text overlap with arXiv:2201.00679

The EMory BrEast imaging Dataset (EMBED): A Racially Diverse, Granular Dataset of 3.5M Screening and Diagnostic Mammograms

Authors: Jiwoong J. Jeong, Brianna L. Vey, Ananth Reddy, Thomas Kim, Thiago Santos, Ramon Correa, Raman Dutt, Marina Mosunjac, Gabriela Oprea-Ilies, Geoffrey Smith, Minjae Woo, Christopher R. McAdams, Mary S. Newell, Imon Banerjee, Judy Gichoya, Hari Trivedi

Abstract: Develo** and validating artificial intelligence models in medical imaging requires datasets that are large, granular, and diverse. To date, the majority of publicly available breast imaging datasets lack in one or more of these areas. Models trained on these data may therefore underperform on patient populations or pathologies that have not previously been encountered. The EMory BrEast imaging D… ▽ More Develo** and validating artificial intelligence models in medical imaging requires datasets that are large, granular, and diverse. To date, the majority of publicly available breast imaging datasets lack in one or more of these areas. Models trained on these data may therefore underperform on patient populations or pathologies that have not previously been encountered. The EMory BrEast imaging Dataset (EMBED) addresses these gaps by providing 3650,000 2D and DBT screening and diagnostic mammograms for 116,000 women divided equally between White and African American patients. The dataset also contains 40,000 annotated lesions linked to structured imaging descriptors and 61 ground truth pathologic outcomes grouped into six severity classes. Our goal is to share this dataset with research partners to aid in development and validation of breast AI models that will serve all patients fairly and help decrease bias in medical AI. △ Less

Submitted 8 February, 2022; originally announced February 2022.

Deciphering signatures of Bardeen black holes from the observed quasi-periodic oscillations

Authors: Indrani Banerjee

Abstract: Quasi-periodic oscillations (QPOs) observed in the power spectrum of black holes are unique observational probes to the background spacetime since they can be directly related to the timescales associated with the motion of matter orbiting in the vicinity of the black hole horizon. In this regard, the high frequency QPOs (HFQPOs) are particularly interesting as they occur in commensurable pairs, t… ▽ More Quasi-periodic oscillations (QPOs) observed in the power spectrum of black holes are unique observational probes to the background spacetime since they can be directly related to the timescales associated with the motion of matter orbiting in the vicinity of the black hole horizon. In this regard, the high frequency QPOs (HFQPOs) are particularly interesting as they occur in commensurable pairs, the most common ratio being the 3:2 twin peak QPOs. The theoretical models which aim to explain these QPOs express the observed frequencies in terms of the epicyclic motion of test particles in a given background spacetime. In this work we study the signatures of Bardeen spacetime from the observed QPOs in the black hole power spectrum. Bardeen black holes are rotating, regular black holes with a magnetic monopole charge. Such regular backgrounds are theoretically interesting as they can potentially evade the curvature singularity, otherwise unavoidable in general relativistic black holes. We perform a chi-square analysis by comparing the available observations of the quasi-periodic oscillations from black hole sources with the relevant theoretical models and note that the Kerr black holes in general relativity are observationally more favored compared to black holes with a monopole charge. Our analysis reveals that black holes with very high monopole charge are disfavored from QPO related observations. △ Less

Submitted 24 September, 2022; v1 submitted 3 January, 2022; originally announced January 2022.

Comments: 23 pages, 2 figures

MedShift: identifying shift data for medical dataset curation

Authors: Xiaoyuan Guo, Judy Wawira Gichoya, Hari Trivedi, Saptarshi Purkayastha, Imon Banerjee

Abstract: To curate a high-quality dataset, identifying data variance between the internal and external sources is a fundamental and crucial step. However, methods to detect shift or variance in data have not been significantly researched. Challenges to this are the lack of effective approaches to learn dense representation of a dataset and difficulties of sharing private data across medical institutions. T… ▽ More To curate a high-quality dataset, identifying data variance between the internal and external sources is a fundamental and crucial step. However, methods to detect shift or variance in data have not been significantly researched. Challenges to this are the lack of effective approaches to learn dense representation of a dataset and difficulties of sharing private data across medical institutions. To overcome the problems, we propose a unified pipeline called MedShift to detect the top-level shift samples and thus facilitate the medical curation. Given an internal dataset A as the base source, we first train anomaly detectors for each class of dataset A to learn internal distributions in an unsupervised way. Second, without exchanging data across sources, we run the trained anomaly detectors on an external dataset B for each class. The data samples with high anomaly scores are identified as shift data. To quantify the shiftness of the external dataset, we cluster B's data into groups class-wise based on the obtained scores. We then train a multi-class classifier on A and measure the shiftness with the classifier's performance variance on B by gradually drop** the group with the largest anomaly score for each class. Additionally, we adapt a dataset quality metric to help inspect the distribution differences for multiple medical sources. We verify the efficacy of MedShift with musculoskeletal radiographs (MURA) and chest X-rays datasets from more than one external source. Experiments show our proposed shift data detection pipeline can be beneficial for medical centers to curate high-quality datasets more efficiently. An interface introduction video to visualize our results is available at https://youtu.be/V3BF0P1sxQE. △ Less

Submitted 27 December, 2021; originally announced December 2021.

Comments: 35 pages, 28 figures, 2 tables

Quasar continuum spectrum disfavors black holes with a magnetic monopole charge

Authors: Indrani Banerjee, Vijay Shersingh Chawan, Bhaswati Mandal, Siddharth Kumar Sahoo, Soumitra SenGupta

Abstract: Black holes carrying a magnetic monopole charge are a subject of interest for a long time. In this work we explore the possibility of an observational evidence of such black holes carrying a magnetic monopole, namely the Bardeen rotating black holes. We derive the theoretical spectrum from the accretion disk surrounding a Bardeen black hole using the thin-disk approximation. We compare the theoret… ▽ More Black holes carrying a magnetic monopole charge are a subject of interest for a long time. In this work we explore the possibility of an observational evidence of such black holes carrying a magnetic monopole, namely the Bardeen rotating black holes. We derive the theoretical spectrum from the accretion disk surrounding a Bardeen black hole using the thin-disk approximation. We compare the theoretically derived spectrum in comparison to the optical data of eighty Palomar Green quasars to constrain the monopole charge parameter $g$ and the spin parameter $a$ of the quasars. From our analysis we note that the Kerr-scenario in \gr\ is observationally more favored than black holes with a monopole charge. We arrive at such a conclusion using error estimators like $χ^2$, the Nash-Sutcliffe efficiency, the index of agreement and their modified forms. In particular, black holes with $g \geq 0.03$ are outside $99\%$ confidence interval. The implications are discussed. △ Less

Submitted 8 February, 2022; v1 submitted 10 December, 2021; originally announced December 2021.

Journal ref: Physical Review D 105, 064073 (2022)

RadFusion: Benchmarking Performance and Fairness for Multimodal Pulmonary Embolism Detection from CT and EHR

Authors: Yuyin Zhou, Shih-Cheng Huang, Jason Alan Fries, Alaa Youssef, Timothy J. Amrhein, Marcello Chang, Imon Banerjee, Daniel Rubin, Lei Xing, Nigam Shah, Matthew P. Lungren

Abstract: Despite the routine use of electronic health record (EHR) data by radiologists to contextualize clinical history and inform image interpretation, the majority of deep learning architectures for medical imaging are unimodal, i.e., they only learn features from pixel-level information. Recent research revealing how race can be recovered from pixel data alone highlights the potential for serious bias… ▽ More Despite the routine use of electronic health record (EHR) data by radiologists to contextualize clinical history and inform image interpretation, the majority of deep learning architectures for medical imaging are unimodal, i.e., they only learn features from pixel-level information. Recent research revealing how race can be recovered from pixel data alone highlights the potential for serious biases in models which fail to account for demographics and other key patient attributes. Yet the lack of imaging datasets which capture clinical context, inclusive of demographics and longitudinal medical history, has left multimodal medical imaging underexplored. To better assess these challenges, we present RadFusion, a multimodal, benchmark dataset of 1794 patients with corresponding EHR data and high-resolution computed tomography (CT) scans labeled for pulmonary embolism. We evaluate several representative multimodal fusion models and benchmark their fairness properties across protected subgroups, e.g., gender, race/ethnicity, age. Our results suggest that integrating imaging and EHR data can improve classification performance and robustness without introducing large disparities in the true positive rate between population groups. △ Less

Submitted 26 November, 2021; v1 submitted 23 November, 2021; originally announced November 2021.

Comments: RadFusion dataset: https://stanfordaimi.azurewebsites.net/datasets/3a7548a4-8f65-4ab7-85fa-3d68c9efc1bd

Two-step adversarial debiasing with partial learning -- medical image case-studies

Authors: Ramon Correa, Jiwoong Jason Jeong, Bhavik Patel, Hari Trivedi, Judy W. Gichoya, Imon Banerjee

Abstract: The use of artificial intelligence (AI) in healthcare has become a very active research area in the last few years. While significant progress has been made in image classification tasks, only a few AI methods are actually being deployed in hospitals. A major hurdle in actively using clinical AI models currently is the trustworthiness of these models. More often than not, these complex models are… ▽ More The use of artificial intelligence (AI) in healthcare has become a very active research area in the last few years. While significant progress has been made in image classification tasks, only a few AI methods are actually being deployed in hospitals. A major hurdle in actively using clinical AI models currently is the trustworthiness of these models. More often than not, these complex models are black boxes in which promising results are generated. However, when scrutinized, these models begin to reveal implicit biases during the decision making, such as detecting race and having bias towards ethnic groups and subpopulations. In our ongoing study, we develop a two-step adversarial debiasing approach with partial learning that can reduce the racial disparity while preserving the performance of the targeted task. The methodology has been evaluated on two independent medical image case-studies - chest X-ray and mammograms, and showed promises in bias reduction while preserving the targeted performance. △ Less

Submitted 16 November, 2021; originally announced November 2021.

CVAD: A generic medical anomaly detector based on Cascade VAE

Authors: Xiaoyuan Guo, Judy Wawira Gichoya, Saptarshi Purkayastha, Imon Banerjee

Abstract: Detecting out-of-distribution (OOD) samples in medical imaging plays an important role for downstream medical diagnosis. However, existing OOD detectors are demonstrated on natural images composed of inter-classes and have difficulty generalizing to medical images. The key issue is the granularity of OOD data in the medical domain, where intra-class OOD samples are predominant. We focus on the gen… ▽ More Detecting out-of-distribution (OOD) samples in medical imaging plays an important role for downstream medical diagnosis. However, existing OOD detectors are demonstrated on natural images composed of inter-classes and have difficulty generalizing to medical images. The key issue is the granularity of OOD data in the medical domain, where intra-class OOD samples are predominant. We focus on the generalizability of OOD detection for medical images and propose a self-supervised Cascade Variational autoencoder-based Anomaly Detector (CVAD). We use a variational autoencoders' cascade architecture, which combines latent representation at multiple scales, before being fed to a discriminator to distinguish the OOD data from the in-distribution (ID) data. Finally, both the reconstruction error and the OOD probability predicted by the binary discriminator are used to determine the anomalies. We compare the performance with the state-of-the-art deep learning models to demonstrate our model's efficacy on various open-access medical imaging datasets for both intra- and inter-class OOD. Further extensive results on datasets including common natural datasets show our model's effectiveness and generalizability. The code is available at https://github.com/XiaoyuanGuo/CVAD. △ Less

Submitted 26 January, 2022; v1 submitted 29 October, 2021; originally announced October 2021.

Comments: 6 pages, 4 figures, 4 tables

Unifying an asymmetric bounce to the dark energy in Chern-Simons F(R) gravity

Authors: Sergei D. Odintsov, Tanmoy Paul, Indrani Banerjee, Ratbay Myrzakulov, Soumitra SenGupta

Abstract: We propose a cosmological scenario in which the universe undergoes through a non-singular bounce, and after the bounce, it decelerates having a matter-like dominated evolution during some regime of the deceleration era, and finally at the present epoch it evolves through an accelerating stage. Our aim is to study such evolution in the context of Chern-Simons corrected F(R) gravity theory and confr… ▽ More We propose a cosmological scenario in which the universe undergoes through a non-singular bounce, and after the bounce, it decelerates having a matter-like dominated evolution during some regime of the deceleration era, and finally at the present epoch it evolves through an accelerating stage. Our aim is to study such evolution in the context of Chern-Simons corrected F(R) gravity theory and confront the model with various observational data. Using the reconstruction technique, and in addition by employing suitable boundary conditions, we determine the form of F(R) for the entire possible range of the cosmic time. The form of F(R) seems to unify a non-singular bounce with a dark energy epoch, in particular, from a non-singular bounce to a deceleration epoch and from a deceleration epoch to a late time acceleration era. It is important to mention that the bouncing scenario in the present context is an asymmetric bounce, in particular, the Hubble radius monotonically increases and asymptotically diverges at the late contracting era, while it seems to decrease with time at the present epoch. Such evolution of the Hubble radius leads to the primordial perturbation modes generate at the deep contracting era when all the perturbation modes lie within the horizon. We calculate the scalar and tensor power spectra, and as a result, the primordial observables are found to be in agreement with the latest Planck 2018 constraints. In this regard, the Chern-Simons term seems to have considerable effects on the tensor perturbation evolution, however kee** intact the scalar part of the perturbation with that of in the case of a vacuum F(R) model, and as a result, the Chern-Simons term proves to play an important role in making the observable quantities consistent with the Planck results. Furthermore the theoretical expectation of the dark energy observables are confronted with the Planck+SNe+BAO data. △ Less

Submitted 31 August, 2021; originally announced September 2021.

Comments: "Physics of the Dark Universe" accepted. Abstract is reduced due to arxiv limitation

Quantum coherence with incomplete set of pointers and corresponding wave-particle duality

Authors: Ingita Banerjee, Kornikar Sen, Chirag Srivastava, Ujjwal Sen

Abstract: Quantum coherence quantifies the amount of superposition in a quantum system, and is the reason and resource behind several phenomena and technologies. It depends on the natural basis in which the quantum state of the system is expressed, which in turn hinges on the physical set-up being analyzed and utilized. While quantum coherence has hitherto been conceptualized by employing different categori… ▽ More Quantum coherence quantifies the amount of superposition in a quantum system, and is the reason and resource behind several phenomena and technologies. It depends on the natural basis in which the quantum state of the system is expressed, which in turn hinges on the physical set-up being analyzed and utilized. While quantum coherence has hitherto been conceptualized by employing different categories of complete bases, there do exist interesting physical situations, where the natural basis is an incomplete one, an example being an interferometric set-up with the observer controlling only a certain fraction of all the slits. We introduce a quantification of quantum coherence with respect to an arbitrary incomplete basis for general quantum states, and develop the corresponding resource theory, identifying the free states and operations. Moreover, we obtain a complementarity relation between the so-defined quantum coherence and the which-path information in an interferometric set-up with several slits, of which only a section is in control of the observer or is accessible to her. This therefore provides us with another face of the wave-particle duality in quantum systems, demonstrating that the complementarity is functional in more general set-ups than thus far considered. △ Less

Submitted 12 August, 2021; originally announced August 2021.

Comments: 10 pages

Analytic topological hairy dyonic black holes and thermodynamics

Authors: Supragyan Priyadarshinee, Subhash Mahapatra, Indrani Banerjee

Abstract: We present and discuss a new family of topological hairy dyonic black hole solutions in asymptotically anti-de Sitter (AdS) space. The coupled Einstein-Maxwell-Scalar gravity system, that carries both the electric and magnetic charges is solved, and exact hairy dyonic black hole solutions are obtained analytically. The scalar field profiles that give rise to such black hole solutions are regular e… ▽ More We present and discuss a new family of topological hairy dyonic black hole solutions in asymptotically anti-de Sitter (AdS) space. The coupled Einstein-Maxwell-Scalar gravity system, that carries both the electric and magnetic charges is solved, and exact hairy dyonic black hole solutions are obtained analytically. The scalar field profiles that give rise to such black hole solutions are regular everywhere. The hairy solutions are obtained for planar, spherical, and hyperbolic horizon topologies. In addition, analytic expressions of regularized action, stress tensor, conserved charges, and free energies are obtained. We further comment on different prescriptions for computing the black hole mass with hairy backgrounds. We analyze the thermodynamics of these hairy dyonic black holes in canonical and grand canonical ensembles, and we find that both electric and magnetic charges have a constructive effect on the stability of the hairy solution. For the case of planar and hyperbolic horizons, we find thermodynamically stable hairy black holes which are favoured at low temperatures compared to the non-hairy counterparts. We further find that, for a spherical hairy dyonic black hole, the thermodynamic phase diagram resembles to that of a Van der Waals fluid not only in canonical but also in the grand canonical ensemble. △ Less

Submitted 6 October, 2021; v1 submitted 5 August, 2021; originally announced August 2021.

Comments: 41 pages, many figures, clarification and references added, published version

Journal ref: Phys. Rev. D 104, 084023 (2021)

A critical analysis of modulus stabilization in a higher dimensional F(R) gravity

Authors: Indrani Banerjee, Tanmoy Paul, Soumitra SenGupta

Abstract: An exact solution for the bulk 5-dimensional geometry is derived for F(R) gravity with non-flat de-Sitter 3-branes located at the $M_4 \times Z_2$ orbifold boundaries. The corresponding form of F(R) that leads to such an exact solution of the bulk metric is derived which turns out to have all positive integer powers of R.In such a scenario the stability issue of the modulus (radion field) is analy… ▽ More An exact solution for the bulk 5-dimensional geometry is derived for F(R) gravity with non-flat de-Sitter 3-branes located at the $M_4 \times Z_2$ orbifold boundaries. The corresponding form of F(R) that leads to such an exact solution of the bulk metric is derived which turns out to have all positive integer powers of R.In such a scenario the stability issue of the modulus (radion field) is analyzed critically for different curvature epochs in both Einstein and Jordan frames. The radion in the effective 4-d theory exhibits a phantom epoch making this model viable for a non-singular bounce. Simultaneous resolution of the gauge hierarchy problem is exhibited through the resulting stable vaue of the radion field in the effective $3+1$ dimensional theory. △ Less

Submitted 4 October, 2021; v1 submitted 1 August, 2021; originally announced August 2021.

Comments: PRD Accepted

Margin-Aware Intra-Class Novelty Identification for Medical Images

Authors: Xiaoyuan Guo, Judy Wawira Gichoya, Saptarshi Purkayastha, Imon Banerjee

Abstract: Traditional anomaly detection methods focus on detecting inter-class variations while medical image novelty identification is inherently an intra-class detection problem. For example, a machine learning model trained with normal chest X-ray and common lung abnormalities, is expected to discover and flag idiopathic pulmonary fibrosis which a rare lung disease and unseen by the model during training… ▽ More Traditional anomaly detection methods focus on detecting inter-class variations while medical image novelty identification is inherently an intra-class detection problem. For example, a machine learning model trained with normal chest X-ray and common lung abnormalities, is expected to discover and flag idiopathic pulmonary fibrosis which a rare lung disease and unseen by the model during training. The nuances from intra-class variations and lack of relevant training data in medical image analysis pose great challenges for existing anomaly detection methods. To tackle the challenges, we propose a hybrid model - Transformation-based Embedding learning for Novelty Detection (TEND) which without any out-of-distribution training data, performs novelty identification by combining both autoencoder-based and classifier-based method. With a pre-trained autoencoder as image feature extractor, TEND learns to discriminate the feature embeddings of in-distribution data from the transformed counterparts as fake out-of-distribution inputs. To enhance the separation, a distance objective is optimized to enforce a margin between the two classes. Extensive experimental results on both natural image datasets and medical image datasets are presented and our method out-performs state-of-the-art approaches. △ Less

Submitted 22 January, 2022; v1 submitted 30 July, 2021; originally announced August 2021.

Comments: 35 pages, 8 figures

Journal ref: Journal of Medical Imaging 2022

Reading Race: AI Recognises Patient's Racial Identity In Medical Images

Authors: Imon Banerjee, Ananth Reddy Bhimireddy, John L. Burns, Leo Anthony Celi, Li-Ching Chen, Ramon Correa, Natalie Dullerud, Marzyeh Ghassemi, Shih-Cheng Huang, Po-Chih Kuo, Matthew P Lungren, Lyle Palmer, Brandon J Price, Saptarshi Purkayastha, Ayis Pyrros, Luke Oakden-Rayner, Chima Okechukwu, Laleh Seyyed-Kalantari, Hari Trivedi, Ryan Wang, Zachary Zaiman, Haoran Zhang, Judy W Gichoya

Abstract: Background: In medical imaging, prior studies have demonstrated disparate AI performance by race, yet there is no known correlation for race on medical imaging that would be obvious to the human expert interpreting the images. Methods: Using private and public datasets we evaluate: A) performance quantification of deep learning models to detect race from medical images, including the ability of… ▽ More Background: In medical imaging, prior studies have demonstrated disparate AI performance by race, yet there is no known correlation for race on medical imaging that would be obvious to the human expert interpreting the images. Methods: Using private and public datasets we evaluate: A) performance quantification of deep learning models to detect race from medical images, including the ability of these models to generalize to external environments and across multiple imaging modalities, B) assessment of possible confounding anatomic and phenotype population features, such as disease distribution and body habitus as predictors of race, and C) investigation into the underlying mechanism by which AI models can recognize race. Findings: Standard deep learning models can be trained to predict race from medical images with high performance across multiple imaging modalities. Our findings hold under external validation conditions, as well as when models are optimized to perform clinically motivated tasks. We demonstrate this detection is not due to trivial proxies or imaging-related surrogate covariates for race, such as underlying disease distribution. Finally, we show that performance persists over all anatomical regions and frequency spectrum of the images suggesting that mitigation efforts will be challenging and demand further study. Interpretation: We emphasize that model ability to predict self-reported race is itself not the issue of importance. However, our findings that AI can trivially predict self-reported race -- even from corrupted, cropped, and noised medical images -- in a setting where clinical experts cannot, creates an enormous risk for all model deployments in medical imaging: if an AI model secretly used its knowledge of self-reported race to misclassify all Black patients, radiologists would not be able to tell using the same data the model has access to. △ Less

Submitted 21 July, 2021; originally announced July 2021.

MSC Class: 68-XX ACM Class: I.2

Looking for extra dimensions in the observed quasi-periodic oscillations of black holes

Authors: Indrani Banerjee, Sumanta Chakraborty, Soumitra SenGupta

Abstract: Quasi-periodic oscillations, often present in the power density spectrum of accretion disk around black holes, are useful probes for the understanding of gravitational interaction in the near-horizon regime of black holes. Since the presence of an extra spatial dimension modifies the near horizon geometry of black holes, it is expected that the study of these quasi-periodic oscillations may shed s… ▽ More Quasi-periodic oscillations, often present in the power density spectrum of accretion disk around black holes, are useful probes for the understanding of gravitational interaction in the near-horizon regime of black holes. Since the presence of an extra spatial dimension modifies the near horizon geometry of black holes, it is expected that the study of these quasi-periodic oscillations may shed some light on the possible existence of these extra dimensions. Intriguingly, most of the extra dimensional models, which are of significant interest to the scientific community, predicts the existence of a tidal charge parameter in black hole spacetimes. This tidal charge parameter can have an overall negative sign and is a distinctive signature of the extra dimensions. Motivated by this, we have studied the quasi-periodic oscillations for a rotating braneworld black hole using the available theoretical models. Subsequently, we have used the observations of the quasi-periodic oscillations from available black hole sources, e.g., GRO J1655 -- 40, XTE J1550 -- 564, GRS 1915 + 105, H 1743 + 322 and Sgr A* and have compared them with the predictions from the relevant theoretical models, in order to estimate the tidal charge parameter. It turns out that among the 11 theoretical models considered here, 8 of them predict a negative value for the tidal charge parameter, while for the others negative values of the tidal charge parameter are also well within the 1-$σ$ confidence interval. △ Less

Submitted 27 September, 2021; v1 submitted 14 May, 2021; originally announced May 2021.

Comments: Published version, 37 pages, 9 figures

Journal ref: JCAP09(2021)037

Dynamic Structural Impact of the COVID-19 Outbreak on the Stock Market and the Exchange Rate: A Cross-country Analysis Among BRICS Nations

Authors: Rupam Bhattacharyya, Sheo Rama, Atul Kumar, Indrajit Banerjee

Abstract: COVID-19 has impacted the economy of almost every country in the world. Of particular interest are the responses of the economic indicators of develo** nations (such as BRICS) to the COVID-19 shock. As an extension to our earlier work on the dynamic associations of pandemic growth, exchange rate, and stock market indices in the context of India, we look at the same question with respect to the B… ▽ More COVID-19 has impacted the economy of almost every country in the world. Of particular interest are the responses of the economic indicators of develo** nations (such as BRICS) to the COVID-19 shock. As an extension to our earlier work on the dynamic associations of pandemic growth, exchange rate, and stock market indices in the context of India, we look at the same question with respect to the BRICS nations. We use structural variable autoregression (SVAR) to identify the dynamic underlying associations across the normalized growth measurements of the COVID-19 cumulative case, recovery, and death counts, and those of the exchange rate, and stock market indices, using data over 203 days (March 12 - September 30, 2020). Using impulse response analyses, the COVID-19 shock to the growth of exchange rate was seen to persist for around 10+ days, and that for stock exchange was seen to be around 15 days. The models capture the contemporaneous nature of these shocks and the subsequent responses, potentially guiding to inform policy decisions at a national level. Further, causal inference-based analyses would allow us to infer relationships that are stronger than mere associations. △ Less

Submitted 10 February, 2021; originally announced February 2021.

Comments: 18 pages, 4 figures, 2 appendices

PAC-Bayes Bounds on Variational Tempered Posteriors for Markov Models

Authors: Imon Banerjee, Vinayak A. Rao, Harsha Honnappa

Abstract: Datasets displaying temporal dependencies abound in science and engineering applications, with Markov models representing a simplified and popular view of the temporal dependence structure. In this paper, we consider Bayesian settings that place prior distributions over the parameters of the transition kernel of a Markov model, and seeks to characterize the resulting, typically intractable, poster… ▽ More Datasets displaying temporal dependencies abound in science and engineering applications, with Markov models representing a simplified and popular view of the temporal dependence structure. In this paper, we consider Bayesian settings that place prior distributions over the parameters of the transition kernel of a Markov model, and seeks to characterize the resulting, typically intractable, posterior distributions. We present a PAC-Bayesian analysis of variational Bayes (VB) approximations to tempered Bayesian posterior distributions, bounding the model risk of the VB approximations. Tempered posteriors are known to be robust to model misspecification, and their variational approximations do not suffer the usual problems of over confident approximations. Our results tie the risk bounds to the mixing and ergodic properties of the Markov data generating model. We illustrate the PAC-Bayes bounds through a number of example Markov models, and also consider the situation where the Markov model is misspecified. △ Less

Submitted 13 January, 2021; originally announced January 2021.

Comments: 14 pages main, 24 pages appendix and citations

Choosing points on cubic plane curves: rigidity and flexibility

Authors: Ishan Banerjee, Weiyan Chen

Abstract: Every smooth cubic plane curve has 9 flex points and 27 sextatic points. We study the following question asked by Farb: Is it true that the known algebraic structures give all the possible ways to continuously choose $n$ distinct points on every smooth cubic plane curve, for each given positive integer $n$? We give an affirmative answer to the question when $n=9$ and 18 (the smallest open cases),… ▽ More Every smooth cubic plane curve has 9 flex points and 27 sextatic points. We study the following question asked by Farb: Is it true that the known algebraic structures give all the possible ways to continuously choose $n$ distinct points on every smooth cubic plane curve, for each given positive integer $n$? We give an affirmative answer to the question when $n=9$ and 18 (the smallest open cases), and a negative answer for infinitely many $n$'s. △ Less

Submitted 24 March, 2021; v1 submitted 11 January, 2021; originally announced January 2021.

Comments: 35 pages, 4 figures. An Appendix was added to the second version, recording a related result that we learned from Curtis McMullen

MSC Class: 55R10 (primary); 55R80; 14H50 (secondary)