Showing 1–29 of 29 results for author: Verma, U

Recovery of synchronized oscillations on multiplex networks by tuning dynamical time scales

Authors: Aiwin T Vadakkan, Umesh Kumar Verma, G. Ambika

Abstract: The heterogeneity among interacting dynamical systems or in the pattern of interactions observed in real complex systems, often lead to partially synchronized states like chimeras or oscillation suppressed states like inhomogeneous or homogeneous steady states. In such cases, recovering synchronized oscillations back is required in many applications but is a real challenge. We present how synchron… ▽ More The heterogeneity among interacting dynamical systems or in the pattern of interactions observed in real complex systems, often lead to partially synchronized states like chimeras or oscillation suppressed states like inhomogeneous or homogeneous steady states. In such cases, recovering synchronized oscillations back is required in many applications but is a real challenge. We present how synchronized oscillations can be restored by tuning the dynamical time scales of the system. For this we use the model of a multiplex network where first layer of coupled oscillators is multiplexed with an environmental layer that can generate various possible chimera states and suppressed states. We show that by tuning the time scale mismatch between the layers , we can revive synchronized oscillations in both layers from these states. We analyse the nature of the transition to synchronization and the results are verified for two- and three-layer multiplex networks. △ Less

Submitted 29 June, 2024; originally announced July 2024.

Comments: 7 pages, 8 figures

Solar Panel Segmentation :Self-Supervised Learning Solutions for Imperfect Datasets

Authors: Sankarshanaa Sagaram, Krish Didwania, Laven Srivastava, Aditya Kasliwal, Pallavi Kailas, Ujjwal Verma

Abstract: The increasing adoption of solar energy necessitates advanced methodologies for monitoring and maintenance to ensure optimal performance of solar panel installations. A critical component in this context is the accurate segmentation of solar panels from aerial or satellite imagery, which is essential for identifying operational issues and assessing efficiency. This paper addresses the significant… ▽ More The increasing adoption of solar energy necessitates advanced methodologies for monitoring and maintenance to ensure optimal performance of solar panel installations. A critical component in this context is the accurate segmentation of solar panels from aerial or satellite imagery, which is essential for identifying operational issues and assessing efficiency. This paper addresses the significant challenges in panel segmentation, particularly the scarcity of annotated data and the labour-intensive nature of manual annotation for supervised learning. We explore and apply Self-Supervised Learning (SSL) to solve these challenges. We demonstrate that SSL significantly enhances model generalization under various conditions and reduces dependency on manually annotated data, paving the way for robust and adaptable solar panel segmentation solutions. △ Less

Submitted 2 June, 2024; v1 submitted 20 February, 2024; originally announced February 2024.

Comments: Published at ICLR Tiny Paper 2024

There Are No Data Like More Data- Datasets for Deep Learning in Earth Observation

Authors: Michael Schmitt, Seyed Ali Ahmadi, Yonghao Xu, Gulsen Taskin, Ujjwal Verma, Francescopaolo Sica, Ronny Hansch

Abstract: Carefully curated and annotated datasets are the foundation of machine learning, with particularly data-hungry deep neural networks forming the core of what is often called Artificial Intelligence (AI). Due to the massive success of deep learning applied to Earth Observation (EO) problems, the focus of the community has been largely on the development of ever-more sophisticated deep neural network… ▽ More Carefully curated and annotated datasets are the foundation of machine learning, with particularly data-hungry deep neural networks forming the core of what is often called Artificial Intelligence (AI). Due to the massive success of deep learning applied to Earth Observation (EO) problems, the focus of the community has been largely on the development of ever-more sophisticated deep neural network architectures and training strategies largely ignoring the overall importance of datasets. For that purpose, numerous task-specific datasets have been created that were largely ignored by previously published review articles on AI for Earth observation. With this article, we want to change the perspective and put machine learning datasets dedicated to Earth observation data and applications into the spotlight. Based on a review of the historical developments, currently available resources are described and a perspective for future developments is formed. We hope to contribute to an understanding that the nature of our data is what distinguishes the Earth observation community from many other communities that apply deep learning techniques to image data, and that a detailed understanding of EO data peculiarities is among the core competencies of our discipline. △ Less

Submitted 29 October, 2023; originally announced October 2023.

Journal ref: Published in IEEE Geoscience and Remote Sensing Magazine, vol. 11, no. 3, pp. 63-97, Sept. 2023

T(w)o patch or not t(w)o patch: A novel additional food model

Authors: Urvashi Verma, Aniket Banerjee, Rana D. Parshad

Abstract: A number of top down bio-control models have been proposed where the predators' efficacy is enhanced via the provision of additional food (AF). These can lead to infinite time \emph{blow-up} of the predator . Predator interference, predator competition, refuge, pest defense and cannibalism have been proposed to achieve pest extinction, while kee** the predator population bounded, but entail \emp… ▽ More A number of top down bio-control models have been proposed where the predators' efficacy is enhanced via the provision of additional food (AF). These can lead to infinite time \emph{blow-up} of the predator . Predator interference, predator competition, refuge, pest defense and cannibalism have been proposed to achieve pest extinction, while kee** the predator population bounded, but entail \emph{quadratic} or higher order dam** terms. Thus an AF model with monotone pest dependent functional response, devoid of these mechanisms, cannot yield the pest extinction state. In the current manuscript we propose a novel patch driven bio-control model, with monotone pest dependent functional response. Here a predator is introduced into a ``patch" such as a prairie strip with AF, and then disperses into a neighboring ``patch" such as a crop field, to target a pest. Assuming only \emph{linear} dispersal in the predator, we show, (i) blow-up in the predator can be prevented in both the prairie strip and the crop field, (ii) the pest extinction state is attainable in the crop field, (iii) the AF model with patch structure can keep pest densities lower than the AF model without patch structure and/or the classical top-down bio-control model, (iv) the AF model with patch structure possesses rich dynamics including chaos. In particular, one observes ``patch specific chaos" - the pest occupying the crop-field can oscillate chaotically, while the pest in the prairie strip oscillates periodically. We discuss these results in light of bio-control strategies that would utilize prairie strips. △ Less

Submitted 25 October, 2023; originally announced October 2023.

Comments: 21 figures, 29 pages

Some Remarks on a Non-Local Variable Carrying Capacity Model for Aphid Population Dynamics

Authors: Aniket Banerjee, Urvashi Verma, Rana D. Parshad

Abstract: Aphids are damaging insect pests on many crops. Their density can rapidly build up on a host plant to several thousand over one growing season. Occasionally, a competition-driven decline in population early in the season, followed by a build-up later, is observed in the field. Such dynamics cannot be captured via standard models, such as introduced in Kindlmann and Dixon, 2010. In Kindlmann et al.… ▽ More Aphids are damaging insect pests on many crops. Their density can rapidly build up on a host plant to several thousand over one growing season. Occasionally, a competition-driven decline in population early in the season, followed by a build-up later, is observed in the field. Such dynamics cannot be captured via standard models, such as introduced in Kindlmann and Dixon, 2010. In Kindlmann et al., 2007, a logistic non-local population model with variable carrying capacity is proposed to capture these alternative dynamics. The proposed model has a rich dynamical structure and can predict multiple population peaks, as observed in the field. We show that additionally, this model also possesses solutions that can blow-up/explode in finite time. The blow-up is seen to occur for both large and small initial conditions and can occur both early and late in the season. We propose a model extension that, under certain parametric restrictions, has global time-bounded solutions. We discuss the ecological applications of our findings. △ Less

Submitted 4 October, 2023; originally announced October 2023.

Comments: 21 pages, 8 figures

Solitary death in coupled limit-cycle oscillators with higher-order interactions

Authors: Subhasanket Dutta, Umesh Kumar Verma, Sarika Jalan

Abstract: Coupled limit cycle oscillators with pairwise interactions depict phase transitions to amplitude or oscillation death. This Letter introduces a scheme for higher-order interactions, which can not be decomposed into pairwise interactions. We investigate Stuart Landau oscillators' dynamical evolution under the impression of such a coupling scheme and discover a particular type of oscillator death wh… ▽ More Coupled limit cycle oscillators with pairwise interactions depict phase transitions to amplitude or oscillation death. This Letter introduces a scheme for higher-order interactions, which can not be decomposed into pairwise interactions. We investigate Stuart Landau oscillators' dynamical evolution under the impression of such a coupling scheme and discover a particular type of oscillator death where a coupling-dependent stable death state, away from the origin, arises in isolation without being accompanied by any other stable state. We call such a state a Solitary death state. Moreover, the explosive transition to the death state is preceded by a surge in amplitude, followed by the revival of the oscillations. Such versatile dynamical states are further enriched with sensitivity to initial conditions. Finally, we point out the resemblance of the results with different dynamical states associated with epileptic seizures. △ Less

Submitted 7 August, 2023; originally announced August 2023.

Explosive death in coupled oscillators with higher-order interactions

Authors: Richita Ghosh, Umesh Kumar Verma, Sarika Jalan, Manish Dev Shrimali

Abstract: We investigate the dynamical evolution of globally connected Stuart-Landau oscillators coupled through conjugate or dis-similar variables on simplicial complexes. We report a first-order explosive phase transition from oscillatory state to death state, with 2-simplex (triadic) interactions, as opposed to the second-order transition with only 1-simplex (dyadic) interactions. Moreover, the system di… ▽ More We investigate the dynamical evolution of globally connected Stuart-Landau oscillators coupled through conjugate or dis-similar variables on simplicial complexes. We report a first-order explosive phase transition from oscillatory state to death state, with 2-simplex (triadic) interactions, as opposed to the second-order transition with only 1-simplex (dyadic) interactions. Moreover, the system displays four distinct homogeneous steady states in the presence of triadic interactions, in contrast to the two homogeneous steady states observed with dyadic interactions. We calculate the backward transition point analytically, confirming the numerical results and providing the origin of the dynamical states in the transition region. The study will be useful in understanding complex systems, such as ecological and epidemiological, having higher-order interactions and coupling through conjugate variables. △ Less

Submitted 7 June, 2023; originally announced June 2023.

Cross-Geography Generalization of Machine Learning Methods for Classification of Flooded Regions in Aerial Images

Authors: Sushant Lenka, Pratyush Kerhalkar, Pranav Shetty, Harsh Gupta, Bhavam Vidyarthi, Ujjwal Verma

Abstract: Identification of regions affected by floods is a crucial piece of information required for better planning and management of post-disaster relief and rescue efforts. Traditionally, remote sensing images are analysed to identify the extent of damage caused by flooding. The data acquired from sensors onboard earth observation satellites are analyzed to detect the flooded regions, which can be affec… ▽ More Identification of regions affected by floods is a crucial piece of information required for better planning and management of post-disaster relief and rescue efforts. Traditionally, remote sensing images are analysed to identify the extent of damage caused by flooding. The data acquired from sensors onboard earth observation satellites are analyzed to detect the flooded regions, which can be affected by low spatial and temporal resolution. However, in recent years, the images acquired from Unmanned Aerial Vehicles (UAVs) have also been utilized to assess post-disaster damage. Indeed, a UAV based platform can be rapidly deployed with a customized flight plan and minimum dependence on the ground infrastructure. This work proposes two approaches for identifying flooded regions in UAV aerial images. The first approach utilizes texture-based unsupervised segmentation to detect flooded areas, while the second uses an artificial neural network on the texture features to classify images as flooded and non-flooded. Unlike the existing works where the models are trained and tested on images of the same geographical regions, this work studies the performance of the proposed model in identifying flooded regions across geographical regions. An F1-score of 0.89 is obtained using the proposed segmentation-based approach which is higher than existing classifiers. The robustness of the proposed approach demonstrates that it can be utilized to identify flooded regions of any region with minimum or no user intervention. △ Less

Submitted 4 October, 2022; originally announced October 2022.

Enhanced Vehicle Re-identification for ITS: A Feature Fusion approach using Deep Learning

Authors: Ashutosh Holla B, Manohara Pai M. M, Ujjwal Verma, Radhika M. Pai

Abstract: In recent years, the development of robust Intelligent transportation systems (ITS) is tackled across the globe to provide better traffic efficiency by reducing frequent traffic problems. As an application of ITS, vehicle re-identification has gained ample interest in the domain of computer vision and robotics. Convolutional neural network (CNN) based methods are developed to perform vehicle re-id… ▽ More In recent years, the development of robust Intelligent transportation systems (ITS) is tackled across the globe to provide better traffic efficiency by reducing frequent traffic problems. As an application of ITS, vehicle re-identification has gained ample interest in the domain of computer vision and robotics. Convolutional neural network (CNN) based methods are developed to perform vehicle re-identification to address key challenges such as occlusion, illumination change, scale, etc. The advancement of transformers in computer vision has opened an opportunity to explore the re-identification process further to enhance performance. In this paper, a framework is developed to perform the re-identification of vehicles across CCTV cameras. To perform re-identification, the proposed framework fuses the vehicle representation learned using a CNN and a transformer model. The framework is tested on a dataset that contains 81 unique vehicle identities observed across 20 CCTV cameras. From the experiments, the fused vehicle re-identification framework yields an mAP of 61.73% which is significantly better when compared with the standalone CNN or transformer model. △ Less

Submitted 13 August, 2022; originally announced August 2022.

Explosive synchronization in coupled nonlinear oscillators on multiplex network

Authors: Umesh Kumar Verma, G. Ambika

Abstract: We report the emergence of explosive synchronization in a multiplex network where oscillators on the first layer are coupled with attractive coupling and those on the second layer, coupled with repulsive coupling. With Stuart-Landau and FitzHugh-Nagumo oscillators as the nodal dynamics, we consider non-local and mean-field intralayer couplings. We establish that explosive synchronization occurs in… ▽ More We report the emergence of explosive synchronization in a multiplex network where oscillators on the first layer are coupled with attractive coupling and those on the second layer, coupled with repulsive coupling. With Stuart-Landau and FitzHugh-Nagumo oscillators as the nodal dynamics, we consider non-local and mean-field intralayer couplings. We establish that explosive synchronization occurs in the multiplex network in the presence of Gaussian noise, and the transition is first order with hysteresis to a state of complete intra-layer and in-phase interlayer synchronization. The width of the hysteresis depends on the range of intralayer attractive coupling, strength of interlayer coupling and noise. We also see how to have control over this induced transition so that the explosive nature, if undesirable, can be converted to continuous type by tuning the strength of repulsive coupling or noise. △ Less

Submitted 6 September, 2022; v1 submitted 11 July, 2022; originally announced July 2022.

Comments: 6 pages, 9 figures

Journal ref: Physics Letters A 450(2022)128391

Texture based Prototypical Network for Few-Shot Semantic Segmentation of Forest Cover: Generalizing for Different Geographical Regions

Authors: Gokul P, Ujjwal Verma

Abstract: Forest plays a vital role in reducing greenhouse gas emissions and mitigating climate change besides maintaining the world's biodiversity. The existing satellite-based forest monitoring system utilizes supervised learning approaches that are limited to a particular region and depend on manually annotated data to identify forest. This work envisages forest identification as a few-shot semantic segm… ▽ More Forest plays a vital role in reducing greenhouse gas emissions and mitigating climate change besides maintaining the world's biodiversity. The existing satellite-based forest monitoring system utilizes supervised learning approaches that are limited to a particular region and depend on manually annotated data to identify forest. This work envisages forest identification as a few-shot semantic segmentation task to achieve generalization across different geographical regions. The proposed few-shot segmentation approach incorporates a texture attention module in the prototypical network to highlight the texture features of the forest. Indeed, the forest exhibits a characteristic texture different from other classes, such as road, water, etc. In this work, the proposed approach is trained for identifying tropical forests of South Asia and adapted to determine the temperate forest of Central Europe with the help of a few (one image for 1-shot) manually annotated support images of the temperate forest. An IoU of 0.62 for forest class (1-way 1-shot) was obtained using the proposed method, which is significantly higher (0.46 for PANet) than the existing few-shot semantic segmentation approach. This result demonstrates that the proposed approach can generalize across geographical regions for forest identification, creating an opportunity to develop a global forest cover identification tool. △ Less

Submitted 2 August, 2022; v1 submitted 29 March, 2022; originally announced March 2022.

Comments: 5 pages, 2 figures, includes additional experiments

Contextual Information Based Anomaly Detection for a Multi-Scene UAV Aerial Videos

Authors: Girisha S, Ujjwal Verma, Manohara Pai M M, Radhika M Pai

Abstract: UAV based surveillance is gaining much interest worldwide due to its extensive applications in monitoring wildlife, urban planning, disaster management, campus security, etc. These videos are analyzed for strange/odd/anomalous patterns which are essential aspects of surveillance. But manual analysis of these videos is tedious and laborious. Hence, the development of computer-aided systems for the… ▽ More UAV based surveillance is gaining much interest worldwide due to its extensive applications in monitoring wildlife, urban planning, disaster management, campus security, etc. These videos are analyzed for strange/odd/anomalous patterns which are essential aspects of surveillance. But manual analysis of these videos is tedious and laborious. Hence, the development of computer-aided systems for the analysis of UAV based surveillance videos is crucial. Despite this interest, in literature, several computer aided systems are developed focusing only on CCTV based surveillance videos. These methods are designed for single scene scenarios and lack contextual knowledge which is required for multi-scene scenarios. Furthermore, the lack of standard UAV based anomaly detection datasets limits the development of these systems. In this regard, the present work aims at the development of a Computer Aided Decision support system to analyse UAV based surveillance videos. A new UAV based multi-scene anomaly detection dataset is developed with frame-level annotations for the development of computer aided systems. It holistically uses contextual, temporal and appearance features for accurate detection of anomalies. Furthermore, a new inference strategy is proposed that utilizes few anomalous samples along with normal samples to identify better decision boundaries. The proposed method is extensively evaluated on the UAV based anomaly detection dataset and performed competitively with respect to state-of-the-art methods. △ Less

Submitted 29 March, 2022; originally announced March 2022.

Evaluating Predictive Uncertainty and Robustness to Distributional Shift Using Real World Data

Authors: Kumud Lakara, Akshat Bhandari, Pratinav Seth, Ujjwal Verma

Abstract: Most machine learning models operate under the assumption that the training, testing and deployment data is independent and identically distributed (i.i.d.). This assumption doesn't generally hold true in a natural setting. Usually, the deployment data is subject to various types of distributional shifts. The magnitude of a model's performance is proportional to this shift in the distribution of t… ▽ More Most machine learning models operate under the assumption that the training, testing and deployment data is independent and identically distributed (i.i.d.). This assumption doesn't generally hold true in a natural setting. Usually, the deployment data is subject to various types of distributional shifts. The magnitude of a model's performance is proportional to this shift in the distribution of the dataset. Thus it becomes necessary to evaluate a model's uncertainty and robustness to distributional shifts to get a realistic estimate of its expected performance on real-world data. Present methods to evaluate uncertainty and model's robustness are lacking and often fail to paint the full picture. Moreover, most analysis so far has primarily focused on classification tasks. In this paper, we propose more insightful metrics for general regression tasks using the Shifts Weather Prediction Dataset. We also present an evaluation of the baseline methods using these metrics. △ Less

Submitted 12 December, 2021; v1 submitted 8 November, 2021; originally announced November 2021.

Comments: 6 pages, 3 figures, 4 tables

Unraveling the Global Teleconnections of Indian Summer Monsoon Clouds: Expedition from CMIP5 to CMIP6

Authors: Ushnanshu Dutta, Anupam Hazra, Hemantkumar S. Chaudhari, Subodh Kumar Saha, Samir Pokhrel, Utkarsh Verma

Abstract: We have analyzed the teleconnection of total cloud fraction (TCF) with global sea surface temperature (SST) in multi-model ensembles (MME) of the fifth and sixth Coupled Model Intercomparison Projects (CMIP5 and CMIP6). CMIP6-MME has a more robust and realistic teleconnection (TCF and global SST) pattern over the extra-tropics (R ~0.43) and North Atlantic (R ~0.39) region, which in turn resulted i… ▽ More We have analyzed the teleconnection of total cloud fraction (TCF) with global sea surface temperature (SST) in multi-model ensembles (MME) of the fifth and sixth Coupled Model Intercomparison Projects (CMIP5 and CMIP6). CMIP6-MME has a more robust and realistic teleconnection (TCF and global SST) pattern over the extra-tropics (R ~0.43) and North Atlantic (R ~0.39) region, which in turn resulted in improvement of rainfall bias over the Asian summer monsoon (ASM) region. CMIP6-MME can better reproduce the mean TCF and have reduced dry (wet) rainfall bias on land (ocean) over the ASM region. CMIP6-MME has improved the biases of seasonal mean rainfall, TCF, and outgoing longwave radiation (OLR) over the Indian Summer Monsoon (ISM) region by ~40%, ~45%, and ~31%, respectively, than CMIP5-MME and demonstrates better spatial correlation with observation/reanalysis. Results establish the credibility of the CMIP6 models and provide a scientific basis for improving the seasonal prediction of ISM. △ Less

Submitted 20 September, 2021; v1 submitted 15 September, 2021; originally announced September 2021.

Comments: 12 pages, 4 main figures, 2 supplementary figures

Emergent Dynamics and Spatio Temporal Patterns on Multiplex Neuronal Networks

Authors: Umesh Kumar Verma, G. Ambika

Abstract: We present a study on the emergence of a variety of spatio temporal patterns among neurons that are connected in a multiplex framework, with neurons on two layers with different functional couplings. With the Hindmarsh-Rose model for the dynamics of single neurons, we analyze the possible patterns of dynamics in each layer separately and report emergent patterns of activity like in-phase synchroni… ▽ More We present a study on the emergence of a variety of spatio temporal patterns among neurons that are connected in a multiplex framework, with neurons on two layers with different functional couplings. With the Hindmarsh-Rose model for the dynamics of single neurons, we analyze the possible patterns of dynamics in each layer separately and report emergent patterns of activity like in-phase synchronized oscillations and amplitude death for excitatory coupling and anti-phase mixed-mode oscillations in multi-clusters with phase regularities when the connections are inhibitory. When they are multiplexed, with neurons of one layer coupled with excitatory synaptic coupling and neurons of the other layer coupled with inhibitory synaptic coupling, we observe transfer or selection of interesting patterns of collective behavior between the layers. While the revival of oscillations occurs in the layer with excitatory coupling, the transition from anti-phase to in-phase and vice versa is observed in the other layer with inhibitory synaptic coupling. We also discuss how theselection of these spatio temporal patterns can be controlled by tuning the intralayer or interlayer coupling strengths or increasing the range of non-local coupling. With one layer having electrical coupling while the other synaptic coupling of excitatory(inhibitory)type, we find in-phase(antiphase) synchronized patterns of activity among neurons in both layers. △ Less

Submitted 24 November, 2021; v1 submitted 14 September, 2021; originally announced September 2021.

Comments: 11 pages, 8 figures

Effect of recirculation zones on the ventilation of a public washroom

Authors: Krishnendu Sinha, Utkarsh Verma, Janani Srree Murallidharan, Vivek Kumar

Abstract: Air-borne transmission can pose a major risk of infection spread in enclosed spaces. Venting the air out using exhaust fans and ducts is a common approach to mitigate the risk. In this work, we study the air flow set up by an exhaust fan in a typical shared washroom that can be a potential hot spot for COVID-19 transmission. The primary focus is on the regions of recirculating flow that can harbor… ▽ More Air-borne transmission can pose a major risk of infection spread in enclosed spaces. Venting the air out using exhaust fans and ducts is a common approach to mitigate the risk. In this work, we study the air flow set up by an exhaust fan in a typical shared washroom that can be a potential hot spot for COVID-19 transmission. The primary focus is on the regions of recirculating flow that can harbor infectious aerosol for much longer than the well-ventilated parts of the room. Computational fluid dynamics is used to obtain the steady state air flow field, and Lagrangian tracking of particles give the spatial and temporal distribution of infectious aerosol in the domain. It is found that the washbasin located next to the door is in a prominent recirculation zone, and particles injected in this region take much longer to be evacuated. The ventilation rate is found to be governed by the air residence time in the recirculation zone, and it is much higher than the time scale based on fully-mixed reactor model of the room. Increasing the fan flow rate can reduce the ventilation time, but cannot eliminate the recirculation zones in the washroom. △ Less

Submitted 24 August, 2021; originally announced August 2021.

BERT based Transformers lead the way in Extraction of Health Information from Social Media

Authors: Sidharth R, Abhiraj Tiwari, Parthivi Choubey, Saisha Kashyap, Sahil Khose, Kumud Lakara, Nishesh Singh, Ujjwal Verma

Abstract: This paper describes our submissions for the Social Media Mining for Health (SMM4H)2021 shared tasks. We participated in 2 tasks:(1) Classification, extraction and normalization of adverse drug effect (ADE) mentions in English tweets (Task-1) and (2) Classification of COVID-19 tweets containing symptoms(Task-6). Our approach for the first task uses the language representation model RoBERTa with a… ▽ More This paper describes our submissions for the Social Media Mining for Health (SMM4H)2021 shared tasks. We participated in 2 tasks:(1) Classification, extraction and normalization of adverse drug effect (ADE) mentions in English tweets (Task-1) and (2) Classification of COVID-19 tweets containing symptoms(Task-6). Our approach for the first task uses the language representation model RoBERTa with a binary classification head. For the second task, we use BERTweet, based on RoBERTa. Fine-tuning is performed on the pre-trained models for both tasks. The models are placed on top of a custom domain-specific processing pipeline. Our system ranked first among all the submissions for subtask-1(a) with an F1-score of 61%. For subtask-1(b), our system obtained an F1-score of 50% with improvements up to +8% F1 over the score averaged across all submissions. The BERTweet model achieved an F1 score of 94% on SMM4H 2021 Task-6. △ Less

Submitted 15 April, 2021; originally announced April 2021.

Comments: 6 pages, 1 figure

Tip** induced by multiplexing on two layer networks

Authors: Umesh Kumar Verma, G. Ambika

Abstract: We report the study of sudden transitions or tip** in a collection of systems induced due to multiplexing with another network of systems. The emergent dynamics of oscillators on one layer can undergo a sudden transition to steady state due to indirect coupling with a shared environment, mean field couplings and conjugate couplings among them. In all these cases, when multiplexed with another se… ▽ More We report the study of sudden transitions or tip** in a collection of systems induced due to multiplexing with another network of systems. The emergent dynamics of oscillators on one layer can undergo a sudden transition to steady state due to indirect coupling with a shared environment, mean field couplings and conjugate couplings among them. In all these cases, when multiplexed with another set of similar systems, the tip** phenomena are induced on the second layer also with a similar pattern of behaviour. We consider van der Pol oscillator as nodal dynamics with various network topologies like scale free and regular networks with local and nonlocal couplings. We also report how the coupling topology influences the nature of transitions on both layers, under multiplexing. △ Less

Submitted 13 May, 2021; v1 submitted 28 February, 2021; originally announced March 2021.

Comments: 16 pages, 11 figures

Journal ref: Eur. Phys. J. Spec. Top. 2021

Analysis of multi-registry kidney exchange program with individual rationality constraints

Authors: Utkarsh Verma, Narayan Rangaraj

Abstract: Kidney exchange programs have been developed to overcome the compatibility challenges for the patient with incompatible donors in kidney transplantation. A registry of such incompatible donor-recipient pairs is created, and compatibility is achieved through the exchange of donors. Single-center kidney exchange registries do not have a significant pool size for achieving the full benefits of the ki… ▽ More Kidney exchange programs have been developed to overcome the compatibility challenges for the patient with incompatible donors in kidney transplantation. A registry of such incompatible donor-recipient pairs is created, and compatibility is achieved through the exchange of donors. Single-center kidney exchange registries do not have a significant pool size for achieving the full benefits of the kidney exchange program. Thus multi-registry exchange transplants seem to be a natural way forward. There are various challenges in multi-registry exchange programs, such as different constraints for each registry, varying bound on cycle lengths, and data sharing among registries. Researchers have proposed different merging mechanisms with a sequential merger, full merger with fair allocations, and full merger without any restrictions. To form a stable multi-registry sharing mechanism, registries should be individually rational both in the long run as well as in the short run. In this paper, an Integer Programming model for a multi-registry exchange program has been proposed with individual rationality constraints for each registry. A simulation study is conducted to compare the benefits of multi-registry programs with individual registry allocations on Indian data, and multiple analyses have been done. It was concluded that registries with lower arrival rates could benefit 7-9\% in terms of the number of transplants and 4-6\% in terms of quality of matches in a typical kidney exchange registry. Registry with more hard to match patients will benefit significantly larger than registries with more easy to match patients. △ Less

Submitted 11 December, 2020; originally announced December 2020.

UVid-Net: Enhanced Semantic Segmentation of UAV Aerial Videos by Embedding Temporal Information

Authors: Girisha S, Ujjwal Verma, Manohara Pai M M, Radhika Pai

Abstract: Semantic segmentation of aerial videos has been extensively used for decision making in monitoring environmental changes, urban planning, and disaster management. The reliability of these decision support systems is dependent on the accuracy of the video semantic segmentation algorithms. The existing CNN based video semantic segmentation methods have enhanced the image semantic segmentation method… ▽ More Semantic segmentation of aerial videos has been extensively used for decision making in monitoring environmental changes, urban planning, and disaster management. The reliability of these decision support systems is dependent on the accuracy of the video semantic segmentation algorithms. The existing CNN based video semantic segmentation methods have enhanced the image semantic segmentation methods by incorporating an additional module such as LSTM or optical flow for computing temporal dynamics of the video which is a computational overhead. The proposed research work modifies the CNN architecture by incorporating temporal information to improve the efficiency of video semantic segmentation. In this work, an enhanced encoder-decoder based CNN architecture (UVid-Net) is proposed for UAV video semantic segmentation. The encoder of the proposed architecture embeds temporal information for temporally consistent labelling. The decoder is enhanced by introducing the feature-refiner module, which aids in accurate localization of the class labels. The proposed UVid-Net architecture for UAV video semantic segmentation is quantitatively evaluated on extended ManipalUAVid dataset. The performance metric mIoU of 0.79 has been observed which is significantly greater than the other state-of-the-art algorithms. Further, the proposed work produced promising results even for the pre-trained model of UVid-Net on urban street scene with fine tuning the final layer on UAV aerial videos. △ Less

Submitted 27 May, 2021; v1 submitted 29 November, 2020; originally announced November 2020.

Comments: Includes additional discussions/results and comparison with SOTA methods. Published in IEEE JSTARS

Journal ref: Published in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 4115-4127, 2021

Weed Density and Distribution Estimation for Precision Agriculture using Semi-Supervised Learning

Authors: Shantam Shorewala, Armaan Ashfaque, Sidharth R, Ujjwal Verma

Abstract: Uncontrolled growth of weeds can severely affect the crop yield and quality. Unrestricted use of herbicide for weed removal alters biodiversity and cause environmental pollution. Instead, identifying weed-infested regions can aid selective chemical treatment of these regions. Advances in analyzing farm images have resulted in solutions to identify weed plants. However, a majority of these approach… ▽ More Uncontrolled growth of weeds can severely affect the crop yield and quality. Unrestricted use of herbicide for weed removal alters biodiversity and cause environmental pollution. Instead, identifying weed-infested regions can aid selective chemical treatment of these regions. Advances in analyzing farm images have resulted in solutions to identify weed plants. However, a majority of these approaches are based on supervised learning methods which requires huge amount of manually annotated images. As a result, these supervised approaches are economically infeasible for the individual farmer because of the wide variety of plant species being cultivated. In this paper, we propose a deep learning-based semi-supervised approach for robust estimation of weed density and distribution across farmlands using only limited color images acquired from autonomous robots. This weed density and distribution can be useful in a site-specific weed management system for selective treatment of infected areas using autonomous robots. In this work, the foreground vegetation pixels containing crops and weeds are first identified using a Convolutional Neural Network (CNN) based unsupervised segmentation. Subsequently, the weed infected regions are identified using a fine-tuned CNN, eliminating the need for designing hand-crafted features. The approach is validated on two datasets of different crop/weed species (1) Crop Weed Field Image Dataset (CWFID), which consists of carrot plant images and the (2) Sugar Beets dataset. The proposed method is able to localize weed-infested regions a maximum recall of 0.99 and estimate weed density with a maximum accuracy of 82.13%. Hence, the proposed approach is shown to generalize to different plant species without the need for extensive labeled data. △ Less

Submitted 18 February, 2021; v1 submitted 4 November, 2020; originally announced November 2020.

Comments: Included few details about the dataset. This paper has been accepted in IEEE Access. Citation: S. Shorewala, A. Ashfaque, S. R and U. Verma, "Weed Density and Distribution Estimation for Precision Agriculture using Semi-Supervised Learning," in IEEE Access, doi: 10.1109/ACCESS.2021.3057912

Journal ref: IEEE Access, vol 9, 2021

Long term simulation analysis of deceased donor initiated chains in kidney exchange programs

Authors: Utkarsh Verma, Narayan Rangaraj

Abstract: Kidney exchange programs have been developed to find compatible kidneys for recipients who have incompatible donors. On the other hand, patients who do not have a living donor depend upon deceased donor (DD) donations to get a kidney transplant. Currently, in India a deceased donor donates kidneys directly to a deceased donor wait-list. The idea of initiating an exchange chain starting from a dece… ▽ More Kidney exchange programs have been developed to find compatible kidneys for recipients who have incompatible donors. On the other hand, patients who do not have a living donor depend upon deceased donor (DD) donations to get a kidney transplant. Currently, in India a deceased donor donates kidneys directly to a deceased donor wait-list. The idea of initiating an exchange chain starting from a deceased donor kidney is proposed in a few articles (and recently executed in Italy), but no mathematical formulation has been given for this merger. We have introduced an integer programming formulation that creates deceased donor initiated chains, considering both paired exchange registry and deceased donor allocations simultaneously. There is a possibility of overlap between paired exchange registry and deceased donor wait-list registry data as recipients can register for both registries independently. This has also been addressed in the paper. A long term simulation study is done to analyze the gain of these DD initiated chains over time. It suggests that even with small numbers of deceased donors, these chains can increases the number of potential transplants significantly. Also, waiting time and dropout rates in the merged registry decreases substantially. △ Less

Submitted 10 October, 2020; originally announced October 2020.

Comments: 23 pages, manuscript still in progress

Amplitude chimera and chimera death induced by external dynamical agents

Authors: Umesh Kumar Verma, G. Ambika

Abstract: We report the emergence of stable amplitude chimeras and chimera death in a two-layer network where one layer has an ensemble of identical nonlinear oscillators interacting directly through local coupling and indirectly through dynamic agents that form the second layer. The nonlocality in the interaction among the dynamical agents in the second layer induces different types of chimera related dyna… ▽ More We report the emergence of stable amplitude chimeras and chimera death in a two-layer network where one layer has an ensemble of identical nonlinear oscillators interacting directly through local coupling and indirectly through dynamic agents that form the second layer. The nonlocality in the interaction among the dynamical agents in the second layer induces different types of chimera related dynamical states in the first layer. The amplitude chimeras developed in them are found to be extremely stable, while chimera death states are prevalent for increased coupling strengths. The results presented are for a system of coupled Stuart-Landau oscillators and can in general represent systems with short-range interactions coupled to another set of systems with long range interactions. In this case, by tuning the range of interactions among the oscillators or the coupling strength between the two types of systems, we can control the nature of chimera states and the system can be restored to homogeneous steady states. The dynamic agents interacting nonlocally with long-range interactions can be considered as a dynamic environment or medium interacting with the system. We indicate how the second layer can act as a reinforcement mechanism on the first layer under various possible interactions for desirable effects. △ Less

Submitted 12 June, 2020; v1 submitted 11 January, 2020; originally announced January 2020.

Comments: 11 pages, 9 figures

Journal ref: Chaos 30, 043104 (2020)

Advent of Extreme Events in Predator Populations

Authors: Sudhanshu Shekhar Chaurasia, Umesh Kumar Verma, Sudeshna Sinha

Abstract: We study the dynamics of a ring of patches with vegetation-prey-predator populations, coupled through interactions of the Lotka-Volterra type. We find that the system yields aperiodic, recurrent and rare explosive bursts of predator density in a few isolated spatial patches from time to time. Further, the collective predator biomass also exhibits sudden uncorrelated occurrences of large deviations… ▽ More We study the dynamics of a ring of patches with vegetation-prey-predator populations, coupled through interactions of the Lotka-Volterra type. We find that the system yields aperiodic, recurrent and rare explosive bursts of predator density in a few isolated spatial patches from time to time. Further, the collective predator biomass also exhibits sudden uncorrelated occurrences of large deviations as the coupled system evolves. The maximum value of the predator population in a patch, as well as the maximum value of the predator biomass, increases with coupling strength. These trends are further corroborated by fits to Generalized Extreme Value distributions, where the location and scale factor of the distribution increases markedly with coupling strength, indicating the crucial role of coupling interactions in the generation of extreme events. These results indicate how occurrences of extremely large predator populations can emerge in coupled population dynamics, and in a more general context they suggest a generic class of deterministic nonlinear systems that can naturally exhibit extreme events. △ Less

Submitted 18 November, 2019; originally announced November 2019.

Explosive death in nonlinear oscillators coupled by quorum sensing

Authors: Umesh Kumar Verma, Sudhanshu Shekhar Chaurasia, Sudeshna Sinha

Abstract: Many biological and chemical systems exhibit collective behavior in response to the change in their population density. These elements or cells communicate with each other via dynamical agents or signaling molecules. In this work, we explore the dynamics of nonlinear oscillators, specifically Stuart-Landau oscillators and Rayleigh oscillators, interacting globally through dynamical agents in the s… ▽ More Many biological and chemical systems exhibit collective behavior in response to the change in their population density. These elements or cells communicate with each other via dynamical agents or signaling molecules. In this work, we explore the dynamics of nonlinear oscillators, specifically Stuart-Landau oscillators and Rayleigh oscillators, interacting globally through dynamical agents in the surrounding environment modeled as a quorum sensing interaction. The system exhibits the typical continuous second-order transition from oscillatory state to death state, when the oscillation amplitude is small. However, interestingly, when the amplitude of oscillations is large we find that the system shows an abrupt transition from oscillatory to death state, a transition termed "explosive death". So the quorum-sensing form of interaction can induce the usual second-order transition, as well as sudden first-order transitions. Further in case of the explosive death transitions, the oscillatory state and the death state coexist over a range of coupling strengths near the transition point. This emergent regime of hysteresis widens with increasing strength of the mean-field feedback, and is relevant to hysteresis that is widely observed in biological, chemical and physical processes. △ Less

Submitted 24 April, 2019; originally announced April 2019.

Journal ref: Phys. Rev. E 100, 032203 (2019)

IP Solutions for International Kidney Exchange Programmes

Authors: Péter Biró, Márton Gyetvai, Radu Stefan Mincu, Alexandru Popa, Utkarsh Verma

Abstract: In kidney exchange programmes patients with end-stage renal failure may exchange their willing, but incompatible living donors among each other. National kidney exchange programmes are in operation in ten European countries, and some of them have already conducted international exchanges through regulated collaborations. The exchanges are selected by conducting regular matching runs (typically eve… ▽ More In kidney exchange programmes patients with end-stage renal failure may exchange their willing, but incompatible living donors among each other. National kidney exchange programmes are in operation in ten European countries, and some of them have already conducted international exchanges through regulated collaborations. The exchanges are selected by conducting regular matching runs (typically every three months) according to well-defined constraints and optimisation criteria, which may differ across countries. In this work we give integer programming formulations for solving international kidney exchange problems, where the optimisation goals and constraints may be different in the participating countries and various feasibility criteria may apply for the international cycles and chains. We also conduct simulations showing the long-run effects of international collaborations for different pools and under various national restrictions and objectives. △ Less

Submitted 19 April, 2019; v1 submitted 15 April, 2019; originally announced April 2019.

Comments: Journal version submitted to CEJOR journal (VOCAL special issue) of the paper with the same name accepted at the proceedings of VOCAL Optimization Conference 2018

MSC Class: 90C10 ACM Class: G.1.6

Malware Evasion Attack and Defense

Authors: Yonghong Huang, Utkarsh Verma, Celeste Fralick, Gabriel Infante-Lopez, Brajesh Kumarz, Carl Woodward

Abstract: Machine learning (ML) classifiers are vulnerable to adversarial examples. An adversarial example is an input sample which is slightly modified to induce misclassification in an ML classifier. In this work, we investigate white-box and grey-box evasion attacks to an ML-based malware detector and conduct performance evaluations in a real-world setting. We compare the defense approaches in mitigating… ▽ More Machine learning (ML) classifiers are vulnerable to adversarial examples. An adversarial example is an input sample which is slightly modified to induce misclassification in an ML classifier. In this work, we investigate white-box and grey-box evasion attacks to an ML-based malware detector and conduct performance evaluations in a real-world setting. We compare the defense approaches in mitigating the attacks. We propose a framework for deploying grey-box and black-box attacks to malware detection systems. △ Less

Submitted 16 April, 2019; v1 submitted 7 April, 2019; originally announced April 2019.

Comments: Accepted by IEEE DSN-DSML2019

On Optimizing Human-Machine Task Assignments

Authors: Andreas Veit, Michael Wilber, Rajan Vaish, Serge Belongie, James Davis, Vishal Anand, Anshu Aviral, Prithvijit Chakrabarty, Yash Chandak, Sidharth Chaturvedi, Chinmaya Devaraj, Ankit Dhall, Utkarsh Dwivedi, Sanket Gupte, Sharath N. Sridhar, Karthik Paga, Anuj Pahuja, Aditya Raisinghani, Ayush Sharma, Shweta Sharma, Darpana Sinha, Nisarg Thakkar, K. Bala Vignesh, Utkarsh Verma, Kanniganti Abhishek , et al. (26 additional authors not shown)

Abstract: When crowdsourcing systems are used in combination with machine inference systems in the real world, they benefit the most when the machine system is deeply integrated with the crowd workers. However, if researchers wish to integrate the crowd with "off-the-shelf" machine classifiers, this deep integration is not always possible. This work explores two strategies to increase accuracy and decrease… ▽ More When crowdsourcing systems are used in combination with machine inference systems in the real world, they benefit the most when the machine system is deeply integrated with the crowd workers. However, if researchers wish to integrate the crowd with "off-the-shelf" machine classifiers, this deep integration is not always possible. This work explores two strategies to increase accuracy and decrease cost under this setting. First, we show that reordering tasks presented to the human can create a significant accuracy improvement. Further, we show that greedily choosing parameters to maximize machine accuracy is sub-optimal, and joint optimization of the combined system improves performance. △ Less

Submitted 24 September, 2015; originally announced September 2015.

Comments: HCOMP 2015 Work in Progress

Fixed Points for Ciric-G-Contractions in Uniform Spaces Endowed with a Graph

Authors: Aris Aghanians, Kamal Fallahi, Kourosh Nourouzi, Ram U Verma

Abstract: In this paper, we generalize the notion of $λ$-generalized contractions introduced by Ćirić from metric to uniform spaces endowed with a graph and discuss on the existence and uniqueness of fixed points for this type of contractions using the basic entourages. In this paper, we generalize the notion of $λ$-generalized contractions introduced by Ćirić from metric to uniform spaces endowed with a graph and discuss on the existence and uniqueness of fixed points for this type of contractions using the basic entourages. △ Less

Submitted 5 April, 2013; originally announced April 2013.

Comments: 12 pages

MSC Class: 47H10; 05C40