Showing 1–50 of 55 results for author: Tripathy, A

Distributed Matrix-Based Sampling for Graph Neural Network Training

Authors: Alok Tripathy, Katherine Yelick, Aydin Buluc

Abstract: Graph Neural Networks (GNNs) offer a compact and computationally efficient way to learn embeddings and classifications on graph data. GNN models are frequently large, making distributed minibatch training necessary. The primary contribution of this paper is new methods for reducing communication in the sampling step for distributed GNN training. Here, we propose a matrix-based bulk sampling appr… ▽ More Graph Neural Networks (GNNs) offer a compact and computationally efficient way to learn embeddings and classifications on graph data. GNN models are frequently large, making distributed minibatch training necessary. The primary contribution of this paper is new methods for reducing communication in the sampling step for distributed GNN training. Here, we propose a matrix-based bulk sampling approach that expresses sampling as a sparse matrix multiplication (SpGEMM) and samples multiple minibatches at once. When the input graph topology does not fit on a single device, our method distributes the graph and use communication-avoiding SpGEMM algorithms to scale GNN minibatch sampling, enabling GNN training on much larger graphs than those that can fit into a single device memory. When the input graph topology (but not the embeddings) fits in the memory of one GPU, our approach (1) performs sampling without communication, (2) amortizes the overheads of sampling a minibatch, and (3) can represent multiple sampling algorithms by simply using different matrix constructions. In addition to new methods for sampling, we introduce a pipeline that uses our matrix-based bulk sampling approach to provide end-to-end training results. We provide experimental results on the largest Open Graph Benchmark (OGB) datasets on $128$ GPUs, and show that our pipeline is $2.5\times$ faster than Quiver (a distributed extension to PyTorch-Geometric) on a $3$-layer GraphSAGE network. On datasets outside of OGB, we show a $8.46\times$ speedup on $128$ GPUs in per-epoch time. Finally, we show scaling when the graph is distributed across GPUs and scaling for both node-wise and layer-wise sampling algorithms. △ Less

Submitted 19 April, 2024; v1 submitted 6 November, 2023; originally announced November 2023.

Comments: Proceedings of Machine Learning and Systems

Multiple exciton generation in VO2

Authors: S. R. Sahu, S. Khan, A. Tripathy, K. Dey, N. Bano, S. Raj Mohan, M. P. Joshi, S. Verma, B. T. Rao, V. G. Sathe, D. K. Shukla

Abstract: Multiple exciton generation (MEG) is a widely studied phenomenon in semiconductor nanocrystals and quantum dots, aimed at improving the energy conversion efficiency of solar cells. MEG is the process wherein incident photon energy is significantly larger than the band gap, and the resulting photoexcited carriers relax by generating additional electron-hole pairs, rather than decaying by heat dissi… ▽ More Multiple exciton generation (MEG) is a widely studied phenomenon in semiconductor nanocrystals and quantum dots, aimed at improving the energy conversion efficiency of solar cells. MEG is the process wherein incident photon energy is significantly larger than the band gap, and the resulting photoexcited carriers relax by generating additional electron-hole pairs, rather than decaying by heat dissipation. Here, we present an experimental demonstration of MEG in a prototype strongly correlated material, VO2, through photocurrent spectroscopy and ultrafast transient reflectivity measurements, both of which are considered the most prominent ways for detecting MEG in working devices. The key result of this paper is the observation of MEG at room temperature (in a correlated insulating phase of VO2), and the estimated threshold for MEG is 3Eg. We demonstrate an escalated photocurrent due to MEG in VO2, and quantum efficiency is found to exceed 100%. Our studies suggest that this phenomenon is a manifestation of expeditious impact ionization due to stronger electron correlations and could be exploited in a large number of strongly correlated materials. △ Less

Submitted 23 October, 2023; originally announced October 2023.

Comments: 6 pages, 5 figures, Physical Review B

Journal ref: Physical Review B 108, 125133 (2023)

Condensate droplet roaming on nanostructured superhydrophobic surfaces

Authors: Cheuk Wing Edmond Lam, Kartik Regulagadda, Matteo Donati, Abinash Tripathy, Gopal Chandra Pal, Chander Shekhar Sharma, Athanasios Milionis, Dimos Poulikakos

Abstract: Jum** of coalescing condensate droplets from superhydrophobic surfaces is an interesting phenomenon which yields marked heat transfer enhancement over the more explored gravity-driven droplet removal mode in surface condensation, a phase change process of central interest to applications ranging from energy to water harvesting. However, when condensate microdroplets coalesce, they can also spont… ▽ More Jum** of coalescing condensate droplets from superhydrophobic surfaces is an interesting phenomenon which yields marked heat transfer enhancement over the more explored gravity-driven droplet removal mode in surface condensation, a phase change process of central interest to applications ranging from energy to water harvesting. However, when condensate microdroplets coalesce, they can also spontaneously propel themselves omnidirectionally on the surface independent of gravity and grow by feeding from droplets they sweep along the way. Here we observe and explain the physics behind this phenomenon of roaming of coalescing condensate microdroplets on solely nanostructured superhydrophobic surfaces, where the microdroplets are orders of magnitude larger than the underlaying surface nanotexture. We quantify and show that it is the inherent asymmetries in droplet adhesion during condensation, arising from the stochastic nature of nucleation within the nanostructures, that generates the tangential momentum driving the roaming motion. Subsequent dewetting during this conversion initiates a vivid roaming and successive coalescence process, preventing condensate flooding of the surface, and enhancing surface renewal. Finally, we show that the more efficient conversion process of roaming from excess surface energy to kinetic energy results in significantly improved heat transfer efficiency over condensate droplet jum**, the mechanism currently understood as maximum. △ Less

Submitted 17 October, 2023; originally announced October 2023.

Key-phrase boosted unsupervised summary generation for FinTech organization

Authors: Aadit Deshpande, Shreya Goyal, Prateek Nagwanshi, Avinash Tripathy

Abstract: With the recent advances in social media, the use of NLP techniques in social media data analysis has become an emerging research direction. Business organizations can particularly benefit from such an analysis of social media discourse, providing an external perspective on consumer behavior. Some of the NLP applications such as intent detection, sentiment classification, text summarization can he… ▽ More With the recent advances in social media, the use of NLP techniques in social media data analysis has become an emerging research direction. Business organizations can particularly benefit from such an analysis of social media discourse, providing an external perspective on consumer behavior. Some of the NLP applications such as intent detection, sentiment classification, text summarization can help FinTech organizations to utilize the social media language data to find useful external insights and can be further utilized for downstream NLP tasks. Particularly, a summary which highlights the intents and sentiments of the users can be very useful for these organizations to get an external perspective. This external perspective can help organizations to better manage their products, offers, promotional campaigns, etc. However, certain challenges, such as a lack of labeled domain-specific datasets impede further exploration of these tasks in the FinTech domain. To overcome these challenges, we design an unsupervised phrase-based summary generation from social media data, using 'Action-Object' pairs (intent phrases). We evaluated the proposed method with other key-phrase based summary generation methods in the direction of contextual information of various Reddit discussion threads, available in the different summaries. We introduce certain "Context Metrics" such as the number of Unique words, Action-Object pairs, and Noun chunks to evaluate the contextual information retrieved from the source text in these phrase-based summaries. We demonstrate that our methods significantly outperform the baseline on these metrics, thus providing a qualitative and quantitative measure of their efficacy. Proposed framework has been leveraged as a web utility portal hosted within Amex. △ Less

Submitted 16 October, 2023; originally announced October 2023.

Comments: 8 pages, 4 figures

Towards Persistent Memory based Stateful Serverless Computing for Big Data Applications

Authors: Yuze Li, Kevin Assogba, Abhijit Tripathy, Moiz Arif, M. Mustafa Rafique, Ali R. Butt, Dimitrios Nikolopoulos

Abstract: The Function-as-a-service (FaaS) computing model has recently seen significant growth especially for highly scalable, event-driven applications. The easy-to-deploy and cost-efficient fine-grained billing of FaaS is highly attractive to big data applications. However, the stateless nature of serverless platforms poses major challenges when supporting stateful I/O intensive workloads such as a lack… ▽ More The Function-as-a-service (FaaS) computing model has recently seen significant growth especially for highly scalable, event-driven applications. The easy-to-deploy and cost-efficient fine-grained billing of FaaS is highly attractive to big data applications. However, the stateless nature of serverless platforms poses major challenges when supporting stateful I/O intensive workloads such as a lack of native support for stateful execution, state sharing, and inter-function communication. In this paper, we explore the feasibility of performing stateful big data analytics on serverless platforms and improving I/O throughput of functions by using modern storage technologies such as Intel Optane DC Persistent Memory (PMEM). To this end, we propose Marvel, an end-to-end architecture built on top of the popular serverless platform, Apache OpenWhisk and Apache Hadoop. Marvel makes two main contributions: (1) enable stateful function execution on OpenWhisk by maintaining state information in an in-memory caching layer; and (2) provide access to PMEM backed HDFS storage for faster I/O performance. Our evaluation shows that Marvel reduces the overall execution time of big data applications by up to 86.6% compared to current MapReduce implementations on AWS Lambda. △ Less

Submitted 8 September, 2023; v1 submitted 4 September, 2023; originally announced September 2023.

Comments: Not yet ready to be publicly available

Using Geographic Location-based Public Health Features in Survival Analysis

Authors: Navid Seidi, Ardhendu Tripathy, Sajal K. Das

Abstract: Time elapsed till an event of interest is often modeled using the survival analysis methodology, which estimates a survival score based on the input features. There is a resurgence of interest in develo** more accurate prediction models for time-to-event prediction in personalized healthcare using modern tools such as neural networks. Higher quality features and more frequent observations improv… ▽ More Time elapsed till an event of interest is often modeled using the survival analysis methodology, which estimates a survival score based on the input features. There is a resurgence of interest in develo** more accurate prediction models for time-to-event prediction in personalized healthcare using modern tools such as neural networks. Higher quality features and more frequent observations improve the predictions for a patient, however, the impact of including a patient's geographic location-based public health statistics on individual predictions has not been studied. This paper proposes a complementary improvement to survival analysis models by incorporating public health statistics in the input features. We show that including geographic location-based public health information results in a statistically significant improvement in the concordance index evaluated on the Surveillance, Epidemiology, and End Results (SEER) dataset containing nationwide cancer incidence data. The improvement holds for both the standard Cox proportional hazards model and the state-of-the-art Deep Survival Machines model. Our results indicate the utility of geographic location-based public health features in survival analysis. △ Less

Submitted 15 April, 2023; originally announced April 2023.

Journal ref: 2023 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE), 2023, 80-91

Workflows Community Summit 2022: A Roadmap Revolution

Authors: Rafael Ferreira da Silva, Rosa M. Badia, Venkat Bala, Debbie Bard, Peer-Timo Bremer, Ian Buckley, Silvina Caino-Lores, Kyle Chard, Carole Goble, Shantenu Jha, Daniel S. Katz, Daniel Laney, Manish Parashar, Frederic Suter, Nick Tyler, Thomas Uram, Ilkay Altintas, Stefan Andersson, William Arndt, Juan Aznar, Jonathan Bader, Bartosz Balis, Chris Blanton, Kelly Rosa Braghetto, Aharon Brodutch , et al. (80 additional authors not shown)

Abstract: Scientific workflows have become integral tools in broad scientific computing use cases. Science discovery is increasingly dependent on workflows to orchestrate large and complex scientific experiments that range from execution of a cloud-based data preprocessing pipeline to multi-facility instrument-to-edge-to-HPC computational workflows. Given the changing landscape of scientific computing and t… ▽ More Scientific workflows have become integral tools in broad scientific computing use cases. Science discovery is increasingly dependent on workflows to orchestrate large and complex scientific experiments that range from execution of a cloud-based data preprocessing pipeline to multi-facility instrument-to-edge-to-HPC computational workflows. Given the changing landscape of scientific computing and the evolving needs of emerging scientific applications, it is paramount that the development of novel scientific workflows and system functionalities seek to increase the efficiency, resilience, and pervasiveness of existing systems and applications. Specifically, the proliferation of machine learning/artificial intelligence (ML/AI) workflows, need for processing large scale datasets produced by instruments at the edge, intensification of near real-time data processing, support for long-term experiment campaigns, and emergence of quantum computing as an adjunct to HPC, have significantly changed the functional and operational requirements of workflow systems. Workflow systems now need to, for example, support data streams from the edge-to-cloud-to-HPC enable the management of many small-sized files, allow data reduction while ensuring high accuracy, orchestrate distributed services (workflows, instruments, data movement, provenance, publication, etc.) across computing and user facilities, among others. Further, to accelerate science, it is also necessary that these systems implement specifications/standards and APIs for seamless (horizontal and vertical) integration between systems and applications, as well as enabling the publication of workflows and their associated products according to the FAIR principles. This document reports on discussions and findings from the 2022 international edition of the Workflows Community Summit that took place on November 29 and 30, 2022. △ Less

Submitted 31 March, 2023; originally announced April 2023.

Report number: ORNL/TM-2023/2885

Accu-Help: A Machine Learning based Smart Healthcare Framework for Accurate Detection of Obsessive Compulsive Disorder

Authors: Kabita Patel, Ajaya Kumar Tripathy, Laxmi Narayan Padhy, Sujita Kumar Kar, Susanta Kumar Padhy, Saraju Prasad Mohanty

Abstract: In recent years the importance of Smart Healthcare cannot be overstated. The current work proposed to expand the state-of-art of smart healthcare in integrating solutions for Obsessive Compulsive Disorder (OCD). Identification of OCD from oxidative stress biomarkers (OSBs) using machine learning is an important development in the study of OCD. However, this process involves the collection of OCD c… ▽ More In recent years the importance of Smart Healthcare cannot be overstated. The current work proposed to expand the state-of-art of smart healthcare in integrating solutions for Obsessive Compulsive Disorder (OCD). Identification of OCD from oxidative stress biomarkers (OSBs) using machine learning is an important development in the study of OCD. However, this process involves the collection of OCD class labels from hospitals, collection of corresponding OSBs from biochemical laboratories, integrated and labeled dataset creation, use of suitable machine learning algorithm for designing OCD prediction model, and making these prediction models available for different biochemical laboratories for OCD prediction for unlabeled OSBs. Further, from time to time, with significant growth in the volume of the dataset with labeled samples, redesigning the prediction model is required for further use. The whole process requires distributed data collection, data integration, coordination between the hospital and biochemical laboratory, dynamic machine learning OCD prediction mode design using a suitable machine learning algorithm, and making the machine learning model available for the biochemical laboratories. Kee** all these things in mind, Accu-Help a fully automated, smart, and accurate OCD detection conceptual model is proposed to help the biochemical laboratories for efficient detection of OCD from OSBs. OSBs are classified into three classes: Healthy Individual (HI), OCD Affected Individual (OAI), and Genetically Affected Individual (GAI). The main component of this proposed framework is the machine learning OCD prediction model design. In this Accu-Help, a neural network-based approach is presented with an OCD prediction accuracy of 86 percent. △ Less

Submitted 5 December, 2022; originally announced December 2022.

Kinetically Decoupled Electrical and Structural Phase Transitions in VO2

Authors: S. R. Sahu, S. S. Majid, A. Ahad, A. Tripathy, K. Dey, S. Pal, B. K. De, Wen-Pin Hsieh, R. Rawat, V. G. Sathe, D. K. Shukla

Abstract: Vanadium dioxide (VO2) has drawn significant attention for its near room temperature insulator to metal transition and associated structural phase transition. The underlying Physics behind the temperature induced insulator to metal and concomitant structural phase transition in VO2 is yet to be fully understood. We have investigated the kinetics of the above phase transition behaviors of VO2 with… ▽ More Vanadium dioxide (VO2) has drawn significant attention for its near room temperature insulator to metal transition and associated structural phase transition. The underlying Physics behind the temperature induced insulator to metal and concomitant structural phase transition in VO2 is yet to be fully understood. We have investigated the kinetics of the above phase transition behaviors of VO2 with the help of resistivity measurements and Raman spectroscopy. Resistance thermal hysteresis scaling and relaxation measurements across the temperature induced insulator to metal transition reveal the unusual behaviour of this first-order phase transition, whereas Raman relaxation measurements show that the temperature induced structural phase transition in VO2 follows usual behaviour and is consistent with mean field prediction. At higher temperature swee** rates decoupling of insulator to metal transition and structural phase transition have been confirmed. The observed anomalous first order phase transition behavior in VO2 is attributed to the unconventional quasi particle dynamics, i.e. significantly lowered electronic thermal conductivity across insulator to metal transition, which is confirmed by ultrafast optical pump-probe time domain thermoreflectance measurements. △ Less

Submitted 6 October, 2022; originally announced October 2022.

Comments: 25 pages, 8 figures

Monoclinic symmetry at the nanoscale in lead-free ferroelectric BaZr$_{x}$Ti$_{1-x}$O$_{3}$ ceramics

Authors: Koushik Dey, Abinash Tripathy, Shikha Rani Sahu, Himanshu Srivastava, Archna Sagdeo, Joerg Strempfer, Dinesh Kumar Shukla

Abstract: Local structural symmetries play a key role in the functionalities of ferroelectric materials and are often found different from average symmetry. Here, we study the real space nanoscale structure in Pb-free BaZr$_{x}$Ti$_{1-x}$O$_{3}$ (x $\leq$ 0.10) by pair distribution function measurements, complemented by transmission electron microscopy and x-ray diffraction. Our observations show existence… ▽ More Local structural symmetries play a key role in the functionalities of ferroelectric materials and are often found different from average symmetry. Here, we study the real space nanoscale structure in Pb-free BaZr$_{x}$Ti$_{1-x}$O$_{3}$ (x $\leq$ 0.10) by pair distribution function measurements, complemented by transmission electron microscopy and x-ray diffraction. Our observations show existence of the rhombohedrally distorted unit cells; however, at intermediate length scales, at least up to 5 nm, there exist nano-scale correlated regions of monoclinic symmetry. This is complemented by the observation of curved frustrated nanodomains. Further, the average structure is found to have coexisting monoclinic and rhombohedral symmetries. Our observation of a two-phase ferroelectric state is in contrast to interferroelectric instabilities of conventional polymorphic phase boundaries reported for doped BaTiO$_{3}$. △ Less

Submitted 12 May, 2022; originally announced May 2022.

Comments: 7 pages, 6 figures

Coexistence of local structural heterogeneities and long-range ferroelectricity in Pb-free (1-x)Ba(Zr0.2Ti0.8)O3-x(Ba0.7Ca0.3)TiO3 ceramics

Authors: Koushik Dey, Abdul Ahad, Kamini Gautam, Abinash Tripathy, Sofi Suhail Majid, Sonia Francoual, Carsten Richter, MN Singh, Archna Sagdeo, Edmund Welter, Naratip Vittayakorn, Vasant G. Sathe, Rajeev Rawat, Dinesh Kumar Shukla

Abstract: Environmentally benign (1-x)Ba(Ti$_{0.8}$Zr$_{0.2}$)O$_3$-x(Ba$_{0.7}$Ca$_{0.3}$)TiO$_3$ (BZT-BCT) ceramics are promising materials due to their remarkable high piezoresponse [Liu and Ren, Phys. Rev. Lett. \textbf{103}, 257602 (2009)]. In this Letter, by focusing on local and average structure in combination with macroscopic electromechanical and dielectric measurements we demonstrate the structur… ▽ More Environmentally benign (1-x)Ba(Ti$_{0.8}$Zr$_{0.2}$)O$_3$-x(Ba$_{0.7}$Ca$_{0.3}$)TiO$_3$ (BZT-BCT) ceramics are promising materials due to their remarkable high piezoresponse [Liu and Ren, Phys. Rev. Lett. \textbf{103}, 257602 (2009)]. In this Letter, by focusing on local and average structure in combination with macroscopic electromechanical and dielectric measurements we demonstrate the structure property relationship in the tetragonal BZT-BCT ceramic. During high-temperature cubic to tetragonal phase transformation, polar nanoregions are manifested through the spontaneous volume ferroelectrostriction at temperatures below $\sim$ 477 K. Temperature-dependent local structural investigations across the Zr K edge extended x-ray absorption fine structure spectroscopy reveal an anomalous collaboration between the ZrO$_{6}$ and TiO$_6$ octahedra. These octahedra compromise their individuality during polarization development. The presence of domains of submicron size embedded inside the macroscopic ferroelectric regions below T$_{m}$, as well as their hierarchical arrangement, is observed by piezo-response force microscopy. Effects of the existence of the structural/polar heterogeneities below T$_{m}$ are observed also when polarizibilities of the poled and the unpoled samples are compared; the poled sample is found to be more susceptible to the electric field. In addition, by using electric field dependent x-ray diffraction studies we also show that this ceramic under field exhibits reduction of tetragonal distortion, which is consistent with earlier reports. △ Less

Submitted 12 May, 2022; originally announced May 2022.

Comments: 9 pages, 6 figures

Orbifold resolution via hyperkahler quotients: the $D_2$ ALF manifold

Authors: Arnav Tripathy, Max Zimet

Abstract: We propose an infinite-dimensional generalization of Kronheimer's construction of families of hyperkahler manifolds resolving flat orbifold quotients of $\mathbb{R}^4$. As in [Kro89], these manifolds are constructed as hyperkahler quotients of affine spaces. This leads to a study of \emph{singular equivariant instantons} in various dimensions. In this paper, we study singular equivariant Nahm data… ▽ More We propose an infinite-dimensional generalization of Kronheimer's construction of families of hyperkahler manifolds resolving flat orbifold quotients of $\mathbb{R}^4$. As in [Kro89], these manifolds are constructed as hyperkahler quotients of affine spaces. This leads to a study of \emph{singular equivariant instantons} in various dimensions. In this paper, we study singular equivariant Nahm data to produce the family of $D_2$ asymptotically locally flat (ALF) manifolds as a deformation of the flat orbifold $(\mathbb{R}^3 \times S^1)/Z_2$. We furthermore introduce a notion of stability for Nahm data and prove a Donaldson-Uhlenbeck-Yau type theorem to relate real and complex formulations. We use these results to construct a canonical Ehresmann connection on the family of non-singular $D_2$ ALF manifolds. In the complex formulation, we exhibit explicit relationships between these $D_2$ ALF manifolds and corresponding $A_1$ ALE manifolds. We conjecture analogous constructions and results for general orbifold quotients of $\mathbb{R}^{4-r} \times T^r$ with $2 \le r \le 4$. The case $r = 4$ produces K3 manifolds as hyperkahler quotients. △ Less

Submitted 10 April, 2022; v1 submitted 25 March, 2022; originally announced March 2022.

Comments: 142 pages

Multiple exciton generation and giant external quantum efficiency in VO$_2$

Authors: S. R. Sahu, A. Tripathy, K. Dey, N. Mansuri, V. G. Sathe, D. K. Shukla

Abstract: Multiple exciton generation (MEG) is a widely studied phenomenon in semiconductor nanocrystals and quantum dots wherein photo-excited carriers relax by generating additional electron-hole pairs. Here, we present the first experimental observation of MEG and the same leading to giant external quantum efficiency (EQE) in VO$_2$, a prototype strongly correlated material. By employing a photoexcitatio… ▽ More Multiple exciton generation (MEG) is a widely studied phenomenon in semiconductor nanocrystals and quantum dots wherein photo-excited carriers relax by generating additional electron-hole pairs. Here, we present the first experimental observation of MEG and the same leading to giant external quantum efficiency (EQE) in VO$_2$, a prototype strongly correlated material. By employing a photoexcitation (lamda ~ 488 nm) of ~ 4.2 times the bandgap, EQE in VO$_2$ is enhanced up to ~ 170 % at room temperature. Temperature dependent experiments exhibit the direct relation between MEG and strength of electron correlation and suggest that such a phenomenon could be exploited in large number of strongly correlated materials for high performance solar cell research in near future. △ Less

Submitted 26 October, 2023; v1 submitted 10 February, 2022; originally announced February 2022.

Comments: New version is available at arXiv:2310.14835

Measuring the complexity of micro and nanostructured surfaces

Authors: A. Arapis, V. Constantoudis, D. Kontziampasis, A. Milionis, C. W. E. Lam, A. Tripathy, D. Poulikakos, E. Gogolides

Abstract: Nanostructured surfaces usually exhibit complicated morphologies that cannot be described in terms of Euclidean geometry. Simultaneously, they do not constitute fully random noise fields to be characterized by simple stochastics and probability theory. In most cases, nanomorphologies consist of complicated mixtures of order and randomness, which should be described quantitatively if one aims to co… ▽ More Nanostructured surfaces usually exhibit complicated morphologies that cannot be described in terms of Euclidean geometry. Simultaneously, they do not constitute fully random noise fields to be characterized by simple stochastics and probability theory. In most cases, nanomorphologies consist of complicated mixtures of order and randomness, which should be described quantitatively if one aims to control their fabrication and properties. In this work, inspired by recent developments in complexity theory, we propose a method to measure nanomorphology complexity that is based on the deviation from the average symmetry of surfaces. We present the methodology for its calculation and the validation of its performance, using a series of synthetic surfaces where the proposed complexity measure obtains a maximum value at the most heterogeneous morphologies between the fully ordered and fully random cases. Additionally, we measure the complexity of experimental micro and nanostructured surfaces (polymeric and metallic), and demonstrate the usefulness of the proposed method in quantifying the impact of processing conditions on their morphologies. Finally, we hint on the relationship between the complexity measure and the functional properties of surfaces. △ Less

Submitted 2 February, 2022; originally announced February 2022.

Comments: 13 pages, 13 figures. Materials Today: Proceedings, 2021

Measurement of ion backflow fraction in GEM detectors

Authors: A. Tripathy, P. K Sahu, S. Swain, S. Sahu

Abstract: A systematic study is performed to measure the ion backflow fraction of the GEM detectors. The effects of different voltage configurations and Ar/CO_2 gas mixtures, in ratios of 70:30, 80:20 and 90:10, on positive ion fraction are investigated in detail. Moreover, a comparative study is performed between single and quadruple GEM detectors.The ion current with detector effective gain is measured wi… ▽ More A systematic study is performed to measure the ion backflow fraction of the GEM detectors. The effects of different voltage configurations and Ar/CO_2 gas mixtures, in ratios of 70:30, 80:20 and 90:10, on positive ion fraction are investigated in detail. Moreover, a comparative study is performed between single and quadruple GEM detectors.The ion current with detector effective gain is measured with various field configurations and with three proportions of gas mixtures. The ion backflow fraction for the GEM is substantially reduced with the lower drift field. A minimum ion backflow fraction of 18 % is achieved in the single GEM detector with Ar/CO_2 80:20 gas mixture, however, a minimum ion backflow fraction of 3.5 %, 3.0%, and 3.8 % are obtained for a drift field of 0.1kV/cm with Ar/CO_2 70:30, 80:20 and 90:10 gas mixtures, respectively for quadrupole GEM detector. Similar values of effective gain and ion backflow fraction have been found by calculating the current from pulse height spectrum method, obtained in the Multi Channel Analyser. △ Less

Submitted 30 June, 2021; originally announced June 2021.

Comments: 24 pages, To be published in Nuclear Instruments and Methods in Physics Research A

Potential Idiomatic Expression (PIE)-English: Corpus for Classes of Idioms

Authors: Tosin P. Adewumi, Roshanak Vadoodi, Aparajita Tripathy, Konstantina Nikolaidou, Foteini Liwicki, Marcus Liwicki

Abstract: We present a fairly large, Potential Idiomatic Expression (PIE) dataset for Natural Language Processing (NLP) in English. The challenges with NLP systems with regards to tasks such as Machine Translation (MT), word sense disambiguation (WSD) and information retrieval make it imperative to have a labelled idioms dataset with classes such as it is in this work. To the best of the authors' knowledge,… ▽ More We present a fairly large, Potential Idiomatic Expression (PIE) dataset for Natural Language Processing (NLP) in English. The challenges with NLP systems with regards to tasks such as Machine Translation (MT), word sense disambiguation (WSD) and information retrieval make it imperative to have a labelled idioms dataset with classes such as it is in this work. To the best of the authors' knowledge, this is the first idioms corpus with classes of idioms beyond the literal and the general idioms classification. In particular, the following classes are labelled in the dataset: metaphor, simile, euphemism, parallelism, personification, oxymoron, paradox, hyperbole, irony and literal. We obtain an overall inter-annotator agreement (IAA) score, between two independent annotators, of 88.89%. Many past efforts have been limited in the corpus size and classes of samples but this dataset contains over 20,100 samples with almost 1,200 cases of idioms (with their meanings) from 10 classes (or senses). The corpus may also be extended by researchers to meet specific needs. The corpus has part of speech (PoS) tagging from the NLTK library. Classification experiments performed on the corpus to obtain a baseline and comparison among three common models, including the BERT model, give good results. We also make publicly available the corpus and the relevant codes for working with it for NLP tasks. △ Less

Submitted 23 April, 2022; v1 submitted 25 April, 2021; originally announced May 2021.

Comments: Accepted at the International Conference on Language Resources and Evaluation (LREC) 2022

Ultra-Thin Lubricant-Infused Vertical Graphene Nanoscaffolds for High-Performance Dropwise Condensation

Authors: Abinash Tripathy, Cheuk Wing Edmond Lam, Diana Davila, Matteo Donati, Athanasios Milionis, Chander Shekhar Sharma, Dimos Poulikakos

Abstract: Lubricant-infused surfaces (LIS) are highly efficient in repelling water and constitute a very promising family of materials for condensation processes occurring in a broad range of energy applications. However, the performance of LIS in such processes is limited by the inherent thermal resistance imposed by the thickness of the lubricant and supporting surface structure, as well as by the gradual… ▽ More Lubricant-infused surfaces (LIS) are highly efficient in repelling water and constitute a very promising family of materials for condensation processes occurring in a broad range of energy applications. However, the performance of LIS in such processes is limited by the inherent thermal resistance imposed by the thickness of the lubricant and supporting surface structure, as well as by the gradual depletion of the lubricant over time. Here we present a remarkable, ultra-thin (~70 nm) and conductive LIS architecture, obtained by infusing lubricant into a vertically grown graphene nanoscaffold on copper. The ultra-thin nature of the scaffold, combined with the high in-plane thermal conductivity of graphene, drastically minimize earlier limitations, effectively doubling the heat transfer performance compared to a state-of-the-art CuO LIS surface. We show that the effect of the thermal resistance to the heat transfer performance of a LIS surface, although often overlooked, can be so detrimental that a simple nanostructured CuO surface can outperform a CuO LIS surface, despite film condensation on the former. The present vertical graphene LIS is also found to be resistant to lubricant depletion, maintaining stable dropwise condensation for at least ~7 hours with no significant change of advancing contact angle and contact angle hysteresis. The lubricant consumed by the vertical graphene LIS is 52.6% less than the existing state-of-the-art CuO LIS, making also the fabrication process more economical. △ Less

Submitted 7 April, 2021; originally announced April 2021.

Journal ref: ACS Nano 2021, 15, 9, 14305-14315

Scalable Hash Table for NUMA Systems

Authors: Alok Tripathy, Oded Green

Abstract: Hash tables are used in a plethora of applications, including database operations, DNA sequencing, string searching, and many more. As such, there are many parallelized hash tables targeting multicore, distributed, and accelerator-based systems. We present in this work a multi-GPU hash table implementation that can process keys at a throughput comparable to that of distributed hash tables. Distrib… ▽ More Hash tables are used in a plethora of applications, including database operations, DNA sequencing, string searching, and many more. As such, there are many parallelized hash tables targeting multicore, distributed, and accelerator-based systems. We present in this work a multi-GPU hash table implementation that can process keys at a throughput comparable to that of distributed hash tables. Distributed CPU hash tables have received significantly more attention than GPU-based hash tables. We show that a single node with multiple GPUs offers roughly the same performance as a 500-1,000-core CPU-based cluster. Our algorithm's key component is our use of multiple sparse-graph data structures and binning techniques to build the hash table. As has been shown individually, these components can be written with massive parallelism that is amenable to GPU acceleration. Since we focus on an individual node, we also leverage communication primitives that are typically prohibitive in distributed environments. We show that our new multi-GPU algorithm shares many of the same features of the single GPU algorithm -- thus we have efficient collision management capabilities and can deal with a large number of duplicates. We evaluate our algorithm on two multi-GPU compute nodes: 1) an NVIDIA DGX2 server with 16 GPUs and 2) an IBM Power 9 Processor with 6 NVIDIA GPUs. With 32-bit keys, our implementation processes 8B keys per second, comparable to some 500-1,000-core CPU-based clusters and 4X faster than prior single-GPU implementations. △ Less

Submitted 1 April, 2021; originally announced April 2021.

Nearest Neighbor Search Under Uncertainty

Authors: Blake Mason, Ardhendu Tripathy, Robert Nowak

Abstract: Nearest Neighbor Search (NNS) is a central task in knowledge representation, learning, and reasoning. There is vast literature on efficient algorithms for constructing data structures and performing exact and approximate NNS. This paper studies NNS under Uncertainty (NNSU). Specifically, consider the setting in which an NNS algorithm has access only to a stochastic distance oracle that provides a… ▽ More Nearest Neighbor Search (NNS) is a central task in knowledge representation, learning, and reasoning. There is vast literature on efficient algorithms for constructing data structures and performing exact and approximate NNS. This paper studies NNS under Uncertainty (NNSU). Specifically, consider the setting in which an NNS algorithm has access only to a stochastic distance oracle that provides a noisy, unbiased estimate of the distance between any pair of points, rather than the exact distance. This models many situations of practical importance, including NNS based on human similarity judgements, physical measurements, or fast, randomized approximations to exact distances. A naive approach to NNSU could employ any standard NNS algorithm and repeatedly query and average results from the stochastic oracle (to reduce noise) whenever it needs a pairwise distance. The problem is that a sufficient number of repeated queries is unknown in advance; e.g., a point maybe distant from all but one other point (crude distance estimates suffice) or it may be close to a large number of other points (accurate estimates are necessary). This paper shows how ideas from cover trees and multi-armed bandits can be leveraged to develop an NNSU algorithm that has optimal dependence on the dataset size and the (unknown)geometry of the dataset. △ Less

Submitted 8 March, 2021; originally announced March 2021.

Comments: 22 pages

Chernoff Sampling for Active Testing and Extension to Active Regression

Authors: Subhojyoti Mukherjee, Ardhendu Tripathy, Robert Nowak

Abstract: Active learning can reduce the number of samples needed to perform a hypothesis test and to estimate the parameters of a model. In this paper, we revisit the work of Chernoff that described an asymptotically optimal algorithm for performing a hypothesis test. We obtain a novel sample complexity bound for Chernoff's algorithm, with a non-asymptotic term that characterizes its performance at a fixed… ▽ More Active learning can reduce the number of samples needed to perform a hypothesis test and to estimate the parameters of a model. In this paper, we revisit the work of Chernoff that described an asymptotically optimal algorithm for performing a hypothesis test. We obtain a novel sample complexity bound for Chernoff's algorithm, with a non-asymptotic term that characterizes its performance at a fixed confidence level. We also develop an extension of Chernoff sampling that can be used to estimate the parameters of a wide variety of models and we obtain a non-asymptotic bound on the estimation error. We apply our extension of Chernoff sampling to actively learn neural network models and to estimate parameters in real-data linear and non-linear regression problems, where our approach performs favorably to state-of-the-art methods. △ Less

Submitted 10 March, 2022; v1 submitted 14 December, 2020; originally announced December 2020.

Comments: 47 pages, 9 figures

A plethora of K3 metrics

Authors: Arnav Tripathy, Max Zimet

Abstract: We extend our recent study of K3 metrics near the $T^4/Z_2$ orbifold locus to the other torus orbifold loci. In particular, we provide several new constructions of K3 surfaces as hyper-Kähler quotients, which yield new formulae for K3 metrics. We then relate these to the construction of arXiv:1810.10540. As a corollary, we derive infinitely many constraints on the (as yet unknown) BPS spectra of t… ▽ More We extend our recent study of K3 metrics near the $T^4/Z_2$ orbifold locus to the other torus orbifold loci. In particular, we provide several new constructions of K3 surfaces as hyper-Kähler quotients, which yield new formulae for K3 metrics. We then relate these to the construction of arXiv:1810.10540. As a corollary, we derive infinitely many constraints on the (as yet unknown) BPS spectra of the Minahan-Nemeschansky SCFTs with $E_n$ global symmetry. Specifically, we find linear combinations of $E_n$ characters (evaluated at different points) hiding within K3 metrics and we compute their second order Taylor expansions. We also find novel strong relationships between the BPS spectra of these SCFTs, as well as with that of the $SU(2)$ $N_f = 4$ SCFT. Finally, we provide a new derivation of the class S constructions of these SCFTs and state some experimental observations regarding their BPS spectra. △ Less

Submitted 23 October, 2020; originally announced October 2020.

Comments: 72 pages

Sprayable Thin and Robust Carbon Nanofiber Composite Coating for Extreme Jum** Dropwise Condensation Performance

Authors: Matteo Donati, Cheuk Wing Edmond Lam, Athanasios Milionis, Chander Shekhar Sharma, Abinash Tripathy, Armend Zendeli, Dimos Poulikakos

Abstract: Condensation of water on metallic surfaces is critical for multiple energy conversion processes. Enhancement in condensation heat transfer efficiency often requires surface texturing and hydrophobicity, usually achieved through coatings, to maintain dropwise condensation. However, such surface treatments face conflicting challenges of minimal coating thermal resistance, enhanced coating durability… ▽ More Condensation of water on metallic surfaces is critical for multiple energy conversion processes. Enhancement in condensation heat transfer efficiency often requires surface texturing and hydrophobicity, usually achieved through coatings, to maintain dropwise condensation. However, such surface treatments face conflicting challenges of minimal coating thermal resistance, enhanced coating durability and scalable fabrication. Here we present a thin (~ 2 μm) polytetrafluoroethylene - carbon nanofiber nanocomposite coating which meets these challenges and sustains coalescence-induced jum** droplet condensation for extended periods under highly demanding condensation conditions. Coating durability is achieved through improved substrate adhesion by depositing a sub-micron thick aluminum primer layer. Carbon nanofibers in a polytetrafluoroethylene matrix increase coating thermal conductivity and promote spontaneous surface nano-texturing to achieve superhydrophobicity for condensate microdroplets. The coating material can be deposited through direct spraying, ensuring economical scalability and versatility for a wide range of substrates. We know of no other coating for metallic surfaces that is able to sustain jum** dropwise condensation under shear of steam at 111 degC flowing at ~ 3 m s-1 over the surface for 10 hours and dropwise condensation for an additional 50 hours. Up to ~ 900% improvement in condensation heat transfer coefficient is achieved compared to conventional filmwise condensation. △ Less

Submitted 5 October, 2020; originally announced October 2020.

Journal ref: Adv. Mater. Interfaces 2021, 8, 2001176

Attractors are not algebraic

Authors: Yeuk Hay Joshua Lam, Arnav Tripathy

Abstract: The Attractor Conjecture for Calabi-Yau moduli spaces predicts the algebraicity of the moduli values of certain isolated points picked out by Hodge-theoretic conditions. We provide a family of counterexamples to the Attractor Conjecture in all suitably high, odd dimensions conditional on the Zilber-Pink conjecture. The Attractor Conjecture for Calabi-Yau moduli spaces predicts the algebraicity of the moduli values of certain isolated points picked out by Hodge-theoretic conditions. We provide a family of counterexamples to the Attractor Conjecture in all suitably high, odd dimensions conditional on the Zilber-Pink conjecture. △ Less

Submitted 29 September, 2020; v1 submitted 26 September, 2020; originally announced September 2020.

Comments: Comments welcome! 24 pages

Journal ref: Compositio Math. 160 (2024) 1073-1100

Finding All ε-Good Arms in Stochastic Bandits

Authors: Blake Mason, Lalit Jain, Ardhendu Tripathy, Robert Nowak

Abstract: The pure-exploration problem in stochastic multi-armed bandits aims to find one or more arms with the largest (or near largest) means. Examples include finding an ε-good arm, best-arm identification, top-k arm identification, and finding all arms with means above a specified threshold. However, the problem of finding all ε-good arms has been overlooked in past work, although arguably this may be t… ▽ More The pure-exploration problem in stochastic multi-armed bandits aims to find one or more arms with the largest (or near largest) means. Examples include finding an ε-good arm, best-arm identification, top-k arm identification, and finding all arms with means above a specified threshold. However, the problem of finding all ε-good arms has been overlooked in past work, although arguably this may be the most natural objective in many applications. For example, a virologist may conduct preliminary laboratory experiments on a large candidate set of treatments and move all ε-good treatments into more expensive clinical trials. Since the ultimate clinical efficacy is uncertain, it is important to identify all ε-good candidates. Mathematically, the all-ε-good arm identification problem presents significant new challenges and surprises that do not arise in the pure-exploration objectives studied in the past. We introduce two algorithms to overcome these and demonstrate their great empirical performance on a large-scale crowd-sourced dataset of 2.2M ratings collected by the New Yorker Caption Contest as well as a dataset testing hundreds of possible cancer drugs. △ Less

Submitted 11 September, 2020; v1 submitted 15 June, 2020; originally announced June 2020.

Comments: 93 total pages (8 main pages + appendices), 12 figures, submitted to NeurIPS 2020

K3 metrics

Authors: Shamit Kachru, Arnav Tripathy, Max Zimet

Abstract: We provide an explicit construction of Ricci-flat K3 metrics. It employs the technology of D-geometry, which in the case of interest is equivalent to a hyper-Kähler quotient. We relate it to the construction of arXiv:1810.10540, and in particular show that it contains the solution to the BPS state counting problem (that of computing the BPS index of a heterotic little string theory compactified on… ▽ More We provide an explicit construction of Ricci-flat K3 metrics. It employs the technology of D-geometry, which in the case of interest is equivalent to a hyper-Kähler quotient. We relate it to the construction of arXiv:1810.10540, and in particular show that it contains the solution to the BPS state counting problem (that of computing the BPS index of a heterotic little string theory compactified on $T^2$) discussed therein, which is the data needed for this second construction of K3 metrics. △ Less

Submitted 5 October, 2020; v1 submitted 3 June, 2020; originally announced June 2020.

Comments: 85 pages. V2: corrected typos

Reducing Communication in Graph Neural Network Training

Authors: Alok Tripathy, Katherine Yelick, Aydin Buluc

Abstract: Graph Neural Networks (GNNs) are powerful and flexible neural networks that use the naturally sparse connectivity information of the data. GNNs represent this connectivity as sparse matrices, which have lower arithmetic intensity and thus higher communication costs compared to dense matrices, making GNNs harder to scale to high concurrencies than convolutional or fully-connected neural networks.… ▽ More Graph Neural Networks (GNNs) are powerful and flexible neural networks that use the naturally sparse connectivity information of the data. GNNs represent this connectivity as sparse matrices, which have lower arithmetic intensity and thus higher communication costs compared to dense matrices, making GNNs harder to scale to high concurrencies than convolutional or fully-connected neural networks. We introduce a family of parallel algorithms for training GNNs and show that they can asymptotically reduce communication compared to previous parallel GNN training methods. We implement these algorithms, which are based on 1D, 1.5D, 2D, and 3D sparse-dense matrix multiplication, using torch.distributed on GPU-equipped clusters. Our algorithms optimize communication across the full GNN training pipeline. We train GNNs on over a hundred GPUs on multiple datasets, including a protein network with over a billion edges. △ Less

Submitted 2 September, 2020; v1 submitted 7 May, 2020; originally announced May 2020.

Comments: To appear in International Conference for High Performance Computing, Networking, Storage, and Analysis (SC'20)

Optimal Confidence Regions for the Multinomial Parameter

Authors: Matthew L. Malloy, Ardhendu Tripathy, Robert D. Nowak

Abstract: Construction of tight confidence regions and intervals is central to statistical inference and decision making. This paper develops new theory showing minimum average volume confidence regions for categorical data. More precisely, consider an empirical distribution $\widehat{\boldsymbol{p}}$ generated from $n$ iid realizations of a random variable that takes one of $k$ possible values according to… ▽ More Construction of tight confidence regions and intervals is central to statistical inference and decision making. This paper develops new theory showing minimum average volume confidence regions for categorical data. More precisely, consider an empirical distribution $\widehat{\boldsymbol{p}}$ generated from $n$ iid realizations of a random variable that takes one of $k$ possible values according to an unknown distribution $\boldsymbol{p}$. This is analogous to a single draw from a multinomial distribution. A confidence region is a subset of the probability simplex that depends on $\widehat{\boldsymbol{p}}$ and contains the unknown $\boldsymbol{p}$ with a specified confidence. This paper shows how one can construct minimum average volume confidence regions, answering a long standing question. We also show the optimality of the regions directly translates to optimal confidence intervals of linear functionals such as the mean, implying sample complexity and regret improvements for adaptive machine learning algorithms. △ Less

Submitted 29 January, 2021; v1 submitted 3 February, 2020; originally announced February 2020.

Random magnetic anisotropy driven transitions in layered perovskite LaSrCoO$_4$

Authors: Abdul Ahad, K. Gautam, S. S. Majid, K. Dey, A. Tripathy, F. Rahman, R. J. Choudhary, R. Sankar, A. K. Sinha, S. N. Kaul, D. K. Shukla

Abstract: Attempts to unravel the nature of magnetic ordering in LaSrCoO$_4$ (Co$^{3+}$), a compound intermediate between antiferromagnetic (AFM) La$_2$CoO$_4$ (Co$^{2+}$) and ferromagnetic (FM) Sr$_2$CoO$_4$ (Co$^{4+}$), have met with a limited success so far. In this report, the results of a thorough investigation of dc magnetization and ac susceptibility (ACS) in single-phase LaSrCoO$_4$ provide clinchin… ▽ More Attempts to unravel the nature of magnetic ordering in LaSrCoO$_4$ (Co$^{3+}$), a compound intermediate between antiferromagnetic (AFM) La$_2$CoO$_4$ (Co$^{2+}$) and ferromagnetic (FM) Sr$_2$CoO$_4$ (Co$^{4+}$), have met with a limited success so far. In this report, the results of a thorough investigation of dc magnetization and ac susceptibility (ACS) in single-phase LaSrCoO$_4$ provide clinching evidence for a thermodynamic paramagnetic (PM) - ferromagnetic (FM) phase transition at T$_{c}$ = 220.5 K, followed at lower temperature (T$_{g}$ = 7.7 K) by a transition to the cluster spin glass (CSG) state. Analysis of the low-field Arrott plot isotherms, in the critical region near T$_{c}$, in terms of the Aharony-Pytte scaling equation of state clearly establishes that the PM-FM transition is basically driven by random magnetic anisotropy (RMA). For temperatures below $\approx$ 30 K, large enough RMA destroys long-range FM order by breaking up the infinite FM network into FM clusters of finite size and leads to the formation of a CSG state at temperatures T $\lesssim$ 8 K by promoting freezing of finite FM clusters in random orientations. Increasing strength of the single-ion magnetocrystalline anisotropy (and hence RMA) with decreasing temperature is taken to reflect an increase in the number of low-spin (LS) Co$^{3+}$ ions at the expense of that of high-spin (HS) Co$^{3+}$ ions. At intermediate temperatures (30 K $\lesssim T \lesssim$ 180 K), spin dynamics has contributions from the infinite FM network (fast relaxation governed by a single anisotropy energy barrier) and finite FM clusters (extremely slow stretched exponential relaxation due to hierarchical energy barriers). △ Less

Submitted 16 May, 2023; v1 submitted 31 January, 2020; originally announced February 2020.

Comments: 23 pages, 9 figures

Journal ref: Phys. Rev. B 107, 214405 (2023)

Drop Impact Printing

Authors: Chandantaru Dey Modak, Arvind Kumar, Abinash Tripathy, Prosenjit Sen

Abstract: Hydrodynamic collapse of a central air-cavity during the recoil phase of droplet impact on a superhydrophobic sieve leads to satellite-free generation of a single droplet through the sieve. Two modes of cavity formation and droplet ejection was observed and explained. The volume of the generated droplet scales with the pore size. Based on this phenomenon, we propose a new drop-on-demand printing t… ▽ More Hydrodynamic collapse of a central air-cavity during the recoil phase of droplet impact on a superhydrophobic sieve leads to satellite-free generation of a single droplet through the sieve. Two modes of cavity formation and droplet ejection was observed and explained. The volume of the generated droplet scales with the pore size. Based on this phenomenon, we propose a new drop-on-demand printing technique. Despite significant advancements in inkjet technology, enhancement in mass-loading and particle-size have been limited due to clogging of the printhead nozzle. By replacing the nozzle with a sieve, we demonstrate printing of nanoparticle suspension with 71% mass-loading. Comparatively large particles of 20 micrometer diameter were dispensed in droplets of 80 micrometer diameter. Printing was performed for surface tension as low as 32 mNm-1 and viscosity as high as 33 mPa-s. In comparison to existing techniques, this new way of printing is widely accessible as it is significantly simple and economical. △ Less

Submitted 17 September, 2019; originally announced October 2019.

Journal ref: Nat Commun 11, 4327 (2020)

A de Rham model for complex analytic equivariant elliptic cohomology

Authors: Daniel Berwick-Evans, Arnav Tripathy

Abstract: We construct a cocycle model for complex analytic equivariant elliptic cohomology that refines Grojnowski's theory when the group is connected and Devoto's when the group is finite. We then construct Mathai--Quillen type cocycles for equivariant elliptic Euler and Thom classes, explaining how these are related to positive energy representations of loop groups. Finally, we show that these classes g… ▽ More We construct a cocycle model for complex analytic equivariant elliptic cohomology that refines Grojnowski's theory when the group is connected and Devoto's when the group is finite. We then construct Mathai--Quillen type cocycles for equivariant elliptic Euler and Thom classes, explaining how these are related to positive energy representations of loop groups. Finally, we show that these classes give a unique equivariant refinement of Hopkins' "theorem of the cube" construction of the ${\rm MString}$-orientation of elliptic cohomology. △ Less

Submitted 30 December, 2020; v1 submitted 7 August, 2019; originally announced August 2019.

Comments: Revised based on referee suggestions. To appear in Advances in Mathematics

Semiclassical Entropy of BPS States in 4d $\mathcal{N}=2$ Theories and Counts of Geodesics

Authors: Shamit Kachru, Richard Nally, Arnav Tripathy, Max Zimet

Abstract: We relate a number of results in the theory of flat surfaces to BPS spectra of a class of 4d $\mathcal{N}=2$ supersymmetric quantum field theories arising from M5 branes wrapped on Riemann surfaces -- $A_1$ class S theories. In particular, we apply classic results of Eskin and Masur, which determine the asymptotic growth of geodesic counts at large length on flat surfaces, as well as more recent p… ▽ More We relate a number of results in the theory of flat surfaces to BPS spectra of a class of 4d $\mathcal{N}=2$ supersymmetric quantum field theories arising from M5 branes wrapped on Riemann surfaces -- $A_1$ class S theories. In particular, we apply classic results of Eskin and Masur, which determine the asymptotic growth of geodesic counts at large length on flat surfaces, as well as more recent progress in the mathematics literature, to determine the large mass asymptotics of the BPS spectra of a wide class of such theories at generic points in the Coulomb branch. △ Less

Submitted 24 December, 2019; v1 submitted 27 June, 2019; originally announced June 2019.

MaxGap Bandit: Adaptive Algorithms for Approximate Ranking

Authors: Sumeet Katariya, Ardhendu Tripathy, Robert Nowak

Abstract: This paper studies the problem of adaptively sampling from K distributions (arms) in order to identify the largest gap between any two adjacent means. We call this the MaxGap-bandit problem. This problem arises naturally in approximate ranking, noisy sorting, outlier detection, and top-arm identification in bandits. The key novelty of the MaxGap-bandit problem is that it aims to adaptively determi… ▽ More This paper studies the problem of adaptively sampling from K distributions (arms) in order to identify the largest gap between any two adjacent means. We call this the MaxGap-bandit problem. This problem arises naturally in approximate ranking, noisy sorting, outlier detection, and top-arm identification in bandits. The key novelty of the MaxGap-bandit problem is that it aims to adaptively determine the natural partitioning of the distributions into a subset with larger means and a subset with smaller means, where the split is determined by the largest gap rather than a pre-specified rank or threshold. Estimating an arm's gap requires sampling its neighboring arms in addition to itself, and this dependence results in a novel hardness parameter that characterizes the sample complexity of the problem. We propose elimination and UCB-style algorithms and show that they are minimax optimal. Our experiments show that the UCB-style algorithms require 6-8x fewer samples than non-adaptive sampling to achieve the same error. △ Less

Submitted 2 June, 2019; originally announced June 2019.

Learning Nearest Neighbor Graphs from Noisy Distance Samples

Authors: Blake Mason, Ardhendu Tripathy, Robert Nowak

Abstract: We consider the problem of learning the nearest neighbor graph of a dataset of n items. The metric is unknown, but we can query an oracle to obtain a noisy estimate of the distance between any pair of items. This framework applies to problem domains where one wants to learn people's preferences from responses commonly modeled as noisy distance judgments. In this paper, we propose an active algorit… ▽ More We consider the problem of learning the nearest neighbor graph of a dataset of n items. The metric is unknown, but we can query an oracle to obtain a noisy estimate of the distance between any pair of items. This framework applies to problem domains where one wants to learn people's preferences from responses commonly modeled as noisy distance judgments. In this paper, we propose an active algorithm to find the graph with high probability and analyze its query complexity. In contrast to existing work that forces Euclidean structure, our method is valid for general metrics, assuming only symmetry and the triangle inequality. Furthermore, we demonstrate efficiency of our method empirically and theoretically, needing only O(n log(n)Delta^-2) queries in favorable settings, where Delta^-2 accounts for the effect of noise. Using crowd-sourced data collected for a subset of the UT Zappos50K dataset, we apply our algorithm to learn which shoes people believe are most similar and show that it beats both an active baseline and ordinal embedding. △ Less

Submitted 30 May, 2019; originally announced May 2019.

Comments: 21 total pages (8 main pages + appendices), 7 figures, submitted to NeurIPS 2019

Black holes and Bhargava's invariant theory

Authors: Murat Gunaydin, Shamit Kachru, Arnav Tripathy

Abstract: Attractor black holes in type II string compactifications on $K3 \times T^2$ are in correspondence with equivalence classes of binary quadratic forms. The discriminant of the quadratic form governs the black hole entropy, and the count of attractor black holes at a given entropy is given by a class number. Here, we show this tantalizing relationship between attractors and arithmetic can be general… ▽ More Attractor black holes in type II string compactifications on $K3 \times T^2$ are in correspondence with equivalence classes of binary quadratic forms. The discriminant of the quadratic form governs the black hole entropy, and the count of attractor black holes at a given entropy is given by a class number. Here, we show this tantalizing relationship between attractors and arithmetic can be generalized to a rich family, connecting black holes in supergravity and string models with analogous equivalence classes of more general forms under the action of arithmetic groups. Many of the physical theories involved have played an earlier role in the study of "magical" supergravities, while their mathematical counterparts are directly related to geometry-of-numbers examples in the work of Bhargava et. al. This paper is dedicated to the memory of Peter Freund. The last section is devoted to some of M.G's personal reminiscences of Peter Freund. △ Less

Submitted 27 October, 2020; v1 submitted 6 March, 2019; originally announced March 2019.

Comments: 22 pages; v2: minor changes, references added, reminiscences about Peter Freund added by first author; v3: two footnotes, one note and references added. Typos corrected

K3 metrics from little string theory

Authors: Shamit Kachru, Arnav Tripathy, Max Zimet

Abstract: Certain six-dimensional (1,0) supersymmetric little string theories, when compactified on $T^3$, have moduli spaces of vacua given by smooth K3 surfaces. Using ideas of Gaiotto-Moore-Neitzke, we show that this provides a systematic procedure for determining the Ricci-flat metric on a smooth K3 surface in terms of BPS degeneracies of (compactified) little string theories. Certain six-dimensional (1,0) supersymmetric little string theories, when compactified on $T^3$, have moduli spaces of vacua given by smooth K3 surfaces. Using ideas of Gaiotto-Moore-Neitzke, we show that this provides a systematic procedure for determining the Ricci-flat metric on a smooth K3 surface in terms of BPS degeneracies of (compactified) little string theories. △ Less

Submitted 10 October, 2020; v1 submitted 24 October, 2018; originally announced October 2018.

Comments: 36 pages. v2: added references. v3: updated joke

Recounting Special Lagrangian Cycles in Twistor Families of K3 Surfaces. Or: How I Learned to Stop Worrying and Count BPS States

Authors: Shamit Kachru, Arnav Tripathy, Max Zimet

Abstract: We consider asymptotics of certain BPS state counts in M-theory compactified on a K3 surface. Our investigation is parallel to (and was inspired by) recent work in the mathematics literature by Filip, who studied the asymptotic count of special Lagrangian fibrations of a marked K3 surface, with fibers of volume at most $V_*$, in a generic twistor family of K3 surfaces. We provide an alternate proo… ▽ More We consider asymptotics of certain BPS state counts in M-theory compactified on a K3 surface. Our investigation is parallel to (and was inspired by) recent work in the mathematics literature by Filip, who studied the asymptotic count of special Lagrangian fibrations of a marked K3 surface, with fibers of volume at most $V_*$, in a generic twistor family of K3 surfaces. We provide an alternate proof of Filip's results by adapting tools that Douglas and collaborators have used to count flux vacua and attractor black holes. We similarly relate BPS state counts in 4d ${\cal N}=2$ supersymmetric gauge theories to certain counting problems in billiard dynamics and provide a simple proof of an old result in this field. △ Less

Submitted 13 December, 2019; v1 submitted 26 July, 2018; originally announced July 2018.

Comments: 11 pages, 1 figure

A model for complex analytic equivariant elliptic cohomology from quantum field theory

Authors: Daniel Berwick-Evans, Arnav Tripathy

Abstract: We construct a global geometric model for complex analytic equivariant elliptic cohomology for all compact Lie groups. Cocycles are specified by functions on the space of fields of the two-dimensional sigma model with background gauge fields and $\mathcal{N} = (0, 1)$ supersymmetry. We also consider a theory of free fermions valued in a representation whose partition function is a section of a det… ▽ More We construct a global geometric model for complex analytic equivariant elliptic cohomology for all compact Lie groups. Cocycles are specified by functions on the space of fields of the two-dimensional sigma model with background gauge fields and $\mathcal{N} = (0, 1)$ supersymmetry. We also consider a theory of free fermions valued in a representation whose partition function is a section of a determinant line bundle. We identify this section with a cocycle representative of the (twisted) equivariant elliptic Euler class of the representation. Finally, we show that the moduli stack of $U(1)$-gauge fields carries a multiplication compatible with the complex analytic group structure on the universal (dual) elliptic curve, with the Euler class providing a choice of coordinate. This provides a physical manifestation of the elliptic group law central to the homotopy-theoretic construction of elliptic cohomology. △ Less

Submitted 24 August, 2020; v1 submitted 10 May, 2018; originally announced May 2018.

Comments: Significant updates informed by the de Rham model in 1908.02868

Zero-error Function Computation on a Directed Acyclic Network

Authors: Ardhendu Tripathy, Aditya Ramamoorthy

Abstract: We study the rate region of variable-length source-network codes that are used to compute a function of messages observed over a network. The particular network considered here is the simplest instance of a directed acyclic graph (DAG) that is not a tree. Existing work on zero-error function computation in DAG networks provides bounds on the \textit{computation capacity}, which is a measure of the… ▽ More We study the rate region of variable-length source-network codes that are used to compute a function of messages observed over a network. The particular network considered here is the simplest instance of a directed acyclic graph (DAG) that is not a tree. Existing work on zero-error function computation in DAG networks provides bounds on the \textit{computation capacity}, which is a measure of the amount of communication required per edge in the worst case. This work focuses on the average case: an achievable rate tuple describes the expected amount of communication required on each edge, where the expectation is over the probability mass function of the source messages. We describe a systematic procedure to obtain outer bounds to the rate region for computing an arbitrary demand function at the terminal. Our bounding technique works by lower bounding the entropy of the descriptions observed by the terminal conditioned on the function value and by utilizing the Schur-concave property of the entropy function. We evaluate these bounds for certain example demand functions. △ Less

Submitted 9 May, 2018; originally announced May 2018.

Comments: 18 pages, 2 figures, submitted to IEEE Transactions on Information Theory

Privacy-Preserving Adversarial Networks

Authors: Ardhendu Tripathy, Ye Wang, Prakash Ishwar

Abstract: We propose a data-driven framework for optimizing privacy-preserving data release mechanisms to attain the information-theoretically optimal tradeoff between minimizing distortion of useful data and concealing specific sensitive information. Our approach employs adversarially-trained neural networks to implement randomized mechanisms and to perform a variational approximation of mutual information… ▽ More We propose a data-driven framework for optimizing privacy-preserving data release mechanisms to attain the information-theoretically optimal tradeoff between minimizing distortion of useful data and concealing specific sensitive information. Our approach employs adversarially-trained neural networks to implement randomized mechanisms and to perform a variational approximation of mutual information privacy. We validate our Privacy-Preserving Adversarial Networks (PPAN) framework via proof-of-concept experiments on discrete and continuous synthetic data, as well as the MNIST handwritten digits dataset. For synthetic data, our model-agnostic PPAN approach achieves tradeoff points very close to the optimal tradeoffs that are analytically-derived from model knowledge. In experiments with the MNIST data, we visually demonstrate a learned tradeoff between minimizing the pixel-level distortion versus concealing the written digit. △ Less

Submitted 12 June, 2019; v1 submitted 19 December, 2017; originally announced December 2017.

Comments: 16 pages

MSC Class: 94A15; 68T05; 62B10

Higher genus Siegel forms and multi-center black holes in N=4 supersymmetric string theory

Authors: Frederik Denef, Shamit Kachru, Zimo Sun, Arnav Tripathy

Abstract: We conjecture that the Fourier coefficients of a degree three Siegel form, $1/\sqrt{χ_{18}}$, count the degeneracy of three-center BPS bound states in type II string theory compactified on $K3 \times T^2$. We provide evidence for our conjecture in the form of consistency with physical considerations of wall-crossing, holographic bounds, and the appearance of suitable counting functions (involving… ▽ More We conjecture that the Fourier coefficients of a degree three Siegel form, $1/\sqrt{χ_{18}}$, count the degeneracy of three-center BPS bound states in type II string theory compactified on $K3 \times T^2$. We provide evidence for our conjecture in the form of consistency with physical considerations of wall-crossing, holographic bounds, and the appearance of suitable counting functions (involving the inverse of the modular discriminant $Δ$ and the inverse of the Igusa cusp form $Φ_{10}$) in limits where the count degenerates to involve single-center or two-center objects. △ Less

Submitted 5 December, 2017; originally announced December 2017.

Compressive three-dimensional super-resolution microscopy with speckle-saturated fluorescence excitation

Authors: Marco Pascucci, Sivaramankrishna Ganesan, Aditya Tripathy, Ori Katz, Valentina Emiliani, Marc Guillon

Abstract: Nonlinear structured illumination microscopy (nSIM) is an effective approach for super-resolution wide-field fluorescence microscopy with a theoretically unlimited resolution. In nSIM, carefully designed, highly-contrasted illumination patterns are combined with the saturation of an optical transition to enable sub-diffraction imaging. While the technique proved useful for two-dimensional imaging,… ▽ More Nonlinear structured illumination microscopy (nSIM) is an effective approach for super-resolution wide-field fluorescence microscopy with a theoretically unlimited resolution. In nSIM, carefully designed, highly-contrasted illumination patterns are combined with the saturation of an optical transition to enable sub-diffraction imaging. While the technique proved useful for two-dimensional imaging, extending it to three-dimensions (3D) is challenging due to the fading/fatigue of organic fluorophores under intense cycling conditions. Here, we present a compressed sensing approach that allows for the first time 3D sub-diffraction nSIM of cultured cells by saturating fluorescence excitation. Exploiting the natural orthogonality of transverse speckle illumination planes, 3D probing of the sample is achieved by a single two-dimensional scan. Fluorescence contrast under saturated excitation is ensured by the inherent high density of intensity minima associated with optical vortices in polarized speckle patterns. Compressed speckle microscopy is thus a simple approach that enables 3D super-resolved nSIM imaging with potentially considerably reduced acquisition time and photobleaching.les fast 3D super-resolved imaging with considerably minimized photo-bleaching. △ Less

Submitted 15 October, 2018; v1 submitted 13 October, 2017; originally announced October 2017.

Journal ref: Nat. Commun. 10, 1327 (2019)

BPS jum** loci are automorphic

Authors: Shamit Kachru, Arnav Tripathy

Abstract: We show that BPS jum** loci -- loci in the moduli space of string compactifications where the number of BPS states jumps in an upper semi-continuous manner -- naturally appear as Fourier coefficients of (vector space-valued) automorphic forms. For the case of $T^2$ compactification, the jum** loci are governed by a modular form studied by Hirzebruch and Zagier, while the jum** loci in K3 com… ▽ More We show that BPS jum** loci -- loci in the moduli space of string compactifications where the number of BPS states jumps in an upper semi-continuous manner -- naturally appear as Fourier coefficients of (vector space-valued) automorphic forms. For the case of $T^2$ compactification, the jum** loci are governed by a modular form studied by Hirzebruch and Zagier, while the jum** loci in K3 compactification appear in a story developed by Oda and Kudla-Millson in arithmetic geometry. We also comment on some curious related automorphy in the physics of black hole attractors and flux vacua. △ Less

Submitted 23 July, 2017; v1 submitted 8 June, 2017; originally announced June 2017.

Comments: 22 pages. Comments welcome!

Black Holes and Hurwitz Class Numbers

Authors: Shamit Kachru, Arnav Tripathy

Abstract: We define a natural counting function for BPS black holes in $K3 \times T^2$ compactification of type II string theory, and observe that it is given by a weight 3/2 mock modular form discovered by Zagier. This hints at tantalizing relations connecting black holes, string theory, and number theory. We define a natural counting function for BPS black holes in $K3 \times T^2$ compactification of type II string theory, and observe that it is given by a weight 3/2 mock modular form discovered by Zagier. This hints at tantalizing relations connecting black holes, string theory, and number theory. △ Less

Submitted 17 May, 2017; originally announced May 2017.

Comments: 6 pages, essay awarded third prize in the Gravity Research Foundation 2017 essay competition

Counting spinning dyons in maximal supergravity: The Hodge-elliptic genus for tori

Authors: Nathan Benjamin, Shamit Kachru, Arnav Tripathy

Abstract: We consider $M$-theory compactified on $T^4 \times T^2$ and describe the count of spinning $1/8$-BPS states. This refines the classic count of Maldacena-Moore-Strominger in the physics literature and the recent mathematical work of Bryan-Oberdieck-Pandharipande-Yin, which studied reduced Donaldson-Thomas invariants of abelian surfaces and threefolds. As in previous work on $K3 \times T^2$ compacti… ▽ More We consider $M$-theory compactified on $T^4 \times T^2$ and describe the count of spinning $1/8$-BPS states. This refines the classic count of Maldacena-Moore-Strominger in the physics literature and the recent mathematical work of Bryan-Oberdieck-Pandharipande-Yin, which studied reduced Donaldson-Thomas invariants of abelian surfaces and threefolds. As in previous work on $K3 \times T^2$ compactification, we track angular momenta under both the $SU(2)_L$ and $SU(2)_R$ factors in the 5d little group, providing predictions for the relevant motivic curve counts. △ Less

Submitted 18 April, 2017; originally announced April 2017.

BPS jum** loci and special cycles

Authors: Shamit Kachru, Arnav Tripathy

Abstract: We study BPS jum** loci, or the subloci in moduli spaces of supersymmetric string vacua where BPS states come into existence discontinuously. This phenomenon is distinct from wall-crossing. We argue that these loci should be thought of as special cycles in the sense of Noether-Lefschetz loci or special Shimura subvarieties, which are indeed examples of BPS jum** loci for certain string compact… ▽ More We study BPS jum** loci, or the subloci in moduli spaces of supersymmetric string vacua where BPS states come into existence discontinuously. This phenomenon is distinct from wall-crossing. We argue that these loci should be thought of as special cycles in the sense of Noether-Lefschetz loci or special Shimura subvarieties, which are indeed examples of BPS jum** loci for certain string compactifications. We use the Hodge-elliptic genus as an informative tool, suggesting that our work can be extended to understand the jum** behavior of motivic Donaldson-Thomas invariants. △ Less

Submitted 30 June, 2017; v1 submitted 1 March, 2017; originally announced March 2017.

Comments: 24 pages. Comments welcome!

The hidden symmetry of the heterotic string

Authors: Shamit Kachru, Arnav Tripathy

Abstract: We propose that Borcherds' Fake Monster Lie algebra is a broken symmetry of heterotic string theory compactified on $T^7 \times T^2$. As evidence, we study the fully flavored counting function for BPS instantons contributing to a certain loop amplitude. The result is controlled by $Φ_{12}$, an automorphic form for $O(2, 26, \mathbb{Z})$. The degeneracies it encodes in its Fourier coefficients are… ▽ More We propose that Borcherds' Fake Monster Lie algebra is a broken symmetry of heterotic string theory compactified on $T^7 \times T^2$. As evidence, we study the fully flavored counting function for BPS instantons contributing to a certain loop amplitude. The result is controlled by $Φ_{12}$, an automorphic form for $O(2, 26, \mathbb{Z})$. The degeneracies it encodes in its Fourier coefficients are graded dimensions of a second-quantized Fock space for this large symmetry algebra. This construction provides a concrete realization of Harvey and Moore's proposed relationship between Generalized Kac-Moody symmetries and supersymmetric string vacua. △ Less

Submitted 8 February, 2017; originally announced February 2017.

Comments: Comments welcome!

Sum-networks from undirected graphs: construction and capacity analysis

Authors: Ardhendu Tripathy, Aditya Ramamoorthy

Abstract: We consider a directed acyclic network with multiple sources and multiple terminals where each terminal is interested in decoding the sum of independent sources generated at the source nodes. We describe a procedure whereby a simple undirected graph can be used to construct such a sum-network and demonstrate an upper bound on its computation rate. Furthermore, we show sufficient conditions for the… ▽ More We consider a directed acyclic network with multiple sources and multiple terminals where each terminal is interested in decoding the sum of independent sources generated at the source nodes. We describe a procedure whereby a simple undirected graph can be used to construct such a sum-network and demonstrate an upper bound on its computation rate. Furthermore, we show sufficient conditions for the construction of a linear network code that achieves this upper bound. Our procedure allows us to construct sum-networks that have any arbitrary computation rate $\frac{p}{q}$ (where $p,q$ are non-negative integers). Our work significantly generalizes a previous approach for constructing sum-networks with arbitrary capacities. Specifically, we answer an open question in prior work by demonstrating sum-networks with significantly fewer number of sources and terminals. △ Less

Submitted 22 December, 2016; originally announced December 2016.

Comments: This has an extra Appendix section giving more information about Remark 2 of the original paper published in the 52nd Annual Allerton Conference on Communication, Control and Computing, 2014

A combinatorial divisibility question from noncommutative algebra

Authors: Arnav Tripathy

Abstract: We present a general conjecture on the divisibility of a certain expression in terms of Kostka numbers and their close variants. This conjecture is closely related to a variant of the period-index problem of noncommutative algebra, with partial implications in both directions. We present a description of the connection between these two problems via Schubert calculus as motivation and evidence for… ▽ More We present a general conjecture on the divisibility of a certain expression in terms of Kostka numbers and their close variants. This conjecture is closely related to a variant of the period-index problem of noncommutative algebra, with partial implications in both directions. We present a description of the connection between these two problems via Schubert calculus as motivation and evidence for the conjecture before turning to a proof of the conjecture in a family of cases. △ Less

Submitted 23 November, 2016; originally announced November 2016.

Comments: 14 pages; comments welcome!

Sum-networks from incidence structures: construction and capacity analysis

Authors: Ardhendu Tripathy, Aditya Ramamoorthy

Abstract: A sum-network is an instance of a network coding problem over a directed acyclic network in which each terminal node wants to compute the sum over a finite field of the information observed at all the source nodes. Many characteristics of the well-studied multiple unicast network communication problem also hold for sum-networks due to a known reduction between instances of these two problems. In t… ▽ More A sum-network is an instance of a network coding problem over a directed acyclic network in which each terminal node wants to compute the sum over a finite field of the information observed at all the source nodes. Many characteristics of the well-studied multiple unicast network communication problem also hold for sum-networks due to a known reduction between instances of these two problems. In this work, we describe an algorithm to construct families of sum-network instances using incidence structures. The computation capacity of several of these sum-network families is characterized. We demonstrate that unlike the multiple unicast problem, the computation capacity of sum-networks depends on the characteristic of the finite field over which the sum is computed. This dependence is very strong; we show examples of sum-networks that have a rate-1 solution over one characteristic but a rate close to zero over a different characteristic. Additionally, a sum-network can have an arbitrary different number of computation capacities for different alphabets. This is contrast to the multiple unicast problem where it is known that the capacity is independent of the network coding alphabet. △ Less

Submitted 28 January, 2018; v1 submitted 6 November, 2016; originally announced November 2016.

Comments: Version accepted for publication in IEEE Transactions on Information Theory

The Hodge-elliptic genus, spinning BPS states, and black holes

Authors: Shamit Kachru, Arnav Tripathy

Abstract: We perform a refined count of BPS states in the compactification of M-theory on $K3 \times T^2$, kee** track of the information provided by both the $SU(2)_L$ and $SU(2)_R$ angular momenta in the $SO(4)$ little group. Mathematically, this four variable counting function may be expressed via the motivic Donaldson-Thomas counts of $K3 \times T^2$, simultaneously refining Katz, Klemm, and Pandharip… ▽ More We perform a refined count of BPS states in the compactification of M-theory on $K3 \times T^2$, kee** track of the information provided by both the $SU(2)_L$ and $SU(2)_R$ angular momenta in the $SO(4)$ little group. Mathematically, this four variable counting function may be expressed via the motivic Donaldson-Thomas counts of $K3 \times T^2$, simultaneously refining Katz, Klemm, and Pandharipande's motivic Donaldson-Thomas counts on $K3$ and Oberdieck-Pandharipande's Gromov-Witten counts on $K3 \times T^2$. This provides the first full answer for motivic curve counts of a compact Calabi-Yau threefold. Along the way, we develop a Hodge-elliptic genus for Calabi-Yau manifolds -- a new counting function for BPS states that interpolates between the Hodge polynomial and the elliptic genus of a Calabi-Yau. △ Less

Submitted 9 December, 2016; v1 submitted 7 September, 2016; originally announced September 2016.

Comments: 21 pages. Comments welcome!