arXiv subjects

Abhishek Sharma

Publications and source records attributed to Abhishek Sharma.

At least 109 records · Page 6Linked to original sources

arXiv2020

Neural Conversational QA: Learning to Reason v.s. Exploiting Patterns

Neural Conversational QA tasks like ShARC require systems to answer questions based on the contents of a given passage. On studying recent state-of-the-art models on the ShARCQA task, we found indications that the models learn spurious clues/patterns in the dataset. Furthermore, we show that a heuristic-based program designed to exploit these patterns can have performance comparable to that of the neural models. In this paper we share our findings about four types of patterns found in the ShARC corpus and describe how neural models exploit them. Motivated by the aforementioned findings, we create and share a modified dataset that has fewer spurious patterns, consequently allowing models to learn better.

cs.CL↗

arXiv2020

Geometric Matrix Completion: A Functional View

We propose a totally functional view of geometric matrix completion problem. Differently from existing work, we propose a novel regularization inspired from the functional map literature that is more interpretable and theoretically sound. On synthetic tasks with strong underlying geometric structure, our framework outperforms state of the art by a huge margin (two order of magnitude) demonstrating the potential of our approach. On real datasets, we achieve state-of-the-art results at a fraction of the computational effort of previous methods. Our code is publicly available at https://github.com/Not-IITian/functional-matrix-completion

cs.LG↗

arXiv2020

Weakly Supervised Deep Functional Map for Shape Matching

A variety of deep functional maps have been proposed recently, from fully supervised to totally unsupervised, with a range of loss functions as well as different regularization terms. However, it is still not clear what are minimum ingredients of a deep functional map pipeline and whether such ingredients unify or generalize all recent work on deep functional maps. We show empirically minimum components for obtaining state of the art results with different loss functions, supervised as well as unsupervised. Furthermore, we propose a novel framework designed for both full-to-full as well as partial to full shape matching that achieves state of the art results on several benchmark datasets outperforming even the fully supervised methods by a significant margin. Our code is publicly available at https://github.com/Not-IITian/Weakly-supervised-Functional-map

cs.CV↗

arXiv2020

Communication and networking technologies for UAVs: A survey

With the advancement in drone technology, in just a few years, drones will be assisting humans in every domain. But there are many challenges to be tackled, communication being the chief one. This paper aims at providing insights into the latest UAV (Unmanned Aerial Vehicle) communication technologies through investigation of suitable task modules, antennas, resource handling platforms, and network architectures. Additionally, we explore techniques such as machine learning and path planning to enhance existing drone communication methods. Encryption and optimization techniques for ensuring long lasting and secure communications, as well as for power management, are discussed. Moreover, applications of UAV networks for different contextual uses ranging from navigation to surveillance, URLLC (Ultra reliable and low latency communications), edge computing and work related to artificial intelligence are examined. In particular, the intricate interplay between UAV, advanced cellular communication, and internet of things constitutes one of the focal points of this paper. The survey encompasses lessons learned, insights, challenges, open issues, and future directions in UAV communications. Our literature review reveals the need for more research work on drone to drone and drone to device communications.

eess.SP↗

arXiv2020

Price Optimization in Fashion E-commerce

With the rapid growth in the fashion e-commerce industry, it is becoming extremely challenging for the E-tailers to set an optimal price point for all the products on the platform. By establishing an optimal price point, they can maximize overall revenue and profit for the platform. In this paper, we propose a novel machine learning and optimization technique to find the optimal price point at an individual product level. It comprises three major components. Firstly, we use a demand prediction model to predict the next day demand for each product at a certain discount percentage. Next step, we use the concept of price elasticity of demand to get the multiple demand values by varying the discount percentage. Thus we obtain multiple price demand pairs for each product and we have to choose one of them for the live platform. Typically fashion e-commerce has millions of products, so there can be many permutations. Each permutation will assign a unique price point for all the products, which will sum up to a unique revenue number. To choose the best permutation which gives maximum revenue, a linear programming optimization technique is used. We have deployed the above methods in the live production environment and conducted several AB tests. According to the AB test result, our model is improving the revenue by 1 percent and gross margin by 0.81 percent.

cs.LG↗

arXiv2020

ASAP-NMS: Accelerating Non-Maximum Suppression Using Spatially Aware Priors

The widely adopted sequential variant of Non Maximum Suppression (or Greedy-NMS) is a crucial module for object-detection pipelines. Unfortunately, for the region proposal stage of two/multi-stage detectors, NMS is turning out to be a latency bottleneck due to its sequential nature. In this article, we carefully profile Greedy-NMS iterations to find that a major chunk of computation is wasted in comparing proposals that are already far-away and have a small chance of suppressing each other. We address this issue by comparing only those proposals that are generated from nearby anchors. The translation-invariant property of the anchor lattice affords generation of a lookup table, which provides an efficient access to nearby proposals, during NMS. This leads to an Accelerated NMS algorithm which leverages Spatially Aware Priors, or ASAP-NMS, and improves the latency of the NMS step from 13.6ms to 1.2 ms on a CPU without sacrificing the accuracy of a state-of-the-art two-stage detector on COCO and VOC datasets. Importantly, ASAP-NMS is agnostic to image resolution and can be used as a simple drop-in module during inference. Using ASAP-NMS at run-time only, we obtain an mAP of 44.2\%@25Hz on the COCO dataset with a V100 GPU.

cs.CV↗

arXiv2020

Proposal for a graphene based nano-electro-mechanical reference piezoresistor

Motivated by the recent prediction of anisotropy in piezoresistance of ballistic graphene along longitudinal and transverse directions, we investigate the angular gauge factor of graphene in the ballistic and diffusive regimes using highly efficient quantum transport models. It is shown that the angular guage factor in both ballistic and diffusive graphene between $0^{\circ}$ to $90^{\circ}$ bears a sinusoidal relation with a periodicity of $π$ due to the reduction of six-fold symmetry into a two-fold symmetry as a result of applied strain. The angular gauge factor is zero at critical angles $20^{\circ}$ and $56^{\circ}$ in ballistic and diffusive regimes respectively. Based on these findings, we propose a graphene based ballistic nano-sensor which can be used as a reference piezoresistor in a Wheatstone bridge read-out technique. The reference sensors proposed here are unsusceptible to inherent residual strain present in strain sensors and unwanted strain generated by the vapours in explosives detection. The theoretical models developed in this paper can be applied to explore similar applications in other 2D-Dirac materials. The proposals made here potentially pave the way for implementation of NEMS strain sensors based on the principle of ballistic transport, which will eventually replace MEMS piezoresistance sensors with a decrease in feature size. The presence of strain insenstive ``critical angle'' in graphene may be useful in flexible wearable electronics also.

cond-mat.mes-hall↗

arXiv2020

OpenEDS2020: Open Eyes Dataset

We present the second edition of OpenEDS dataset, OpenEDS2020, a novel dataset of eye-image sequences captured at a frame rate of 100 Hz under controlled illumination, using a virtual-reality head-mounted display mounted with two synchronized eye-facing cameras. The dataset, which is anonymized to remove any personally identifiable information on participants, consists of 80 participants of varied appearance performing several gaze-elicited tasks, and is divided in two subsets: 1) Gaze Prediction Dataset, with up to 66,560 sequences containing 550,400 eye-images and respective gaze vectors, created to foster research in spatio-temporal gaze estimation and prediction approaches; and 2) Eye Segmentation Dataset, consisting of 200 sequences sampled at 5 Hz, with up to 29,500 images, of which 5% contain a semantic segmentation label, devised to encourage the use of temporal information to propagate labels to contiguous frames. Baseline experiments have been evaluated on OpenEDS2020, one for each task, with average angular error of 5.37 degrees when performing gaze prediction on 1 to 5 frames into the future, and a mean intersection over union score of 84.1% for semantic segmentation. As its predecessor, OpenEDS dataset, we anticipate that this new dataset will continue creating opportunities to researchers in eye tracking, machine learning and computer vision communities, to advance the state of the art for virtual reality applications. The dataset is available for download upon request at http://research.fb.com/programs/openeds-2020-challenge/.

cs.CV↗

arXiv2020

Towards Automatic Generation of Questions from Long Answers

Automatic question generation (AQG) has broad applicability in domains such as tutoring systems, conversational agents, healthcare literacy, and information retrieval. Existing efforts at AQG have been limited to short answer lengths of up to two or three sentences. However, several real-world applications require question generation from answers that span several sentences. Therefore, we propose a novel evaluation benchmark to assess the performance of existing AQG systems for long-text answers. We leverage the large-scale open-source Google Natural Questions dataset to create the aforementioned long-answer AQG benchmark. We empirically demonstrate that the performance of existing AQG methods significantly degrades as the length of the answer increases. Transformer-based methods outperform other existing AQG methods on long answers in terms of automatic as well as human evaluation. However, we still observe degradation in the performance of our best performing models with increasing sentence length, suggesting that long answer QA is a challenging benchmark task for future research.

cs.CL↗

arXiv2020

Deep Geometric Functional Maps: Robust Feature Learning for Shape Correspondence

We present a novel learning-based approach for computing correspondences between non-rigid 3D shapes. Unlike previous methods that either require extensive training data or operate on handcrafted input descriptors and thus generalize poorly across diverse datasets, our approach is both accurate and robust to changes in shape structure. Key to our method is a feature-extraction network that learns directly from raw shape geometry, combined with a novel regularized map extraction layer and loss, based on the functional map representation. We demonstrate through extensive experiments in challenging shape matching scenarios that our method can learn from less training data than existing supervised approaches and generalizes significantly better than current descriptor-based learning methods. Our source code is available at: https://github.com/LIX-shape-analysis/GeomFmaps.

stat.ML↗

arXiv2020

Multiview-Consistent Semi-Supervised Learning for 3D Human Pose Estimation

The best performing methods for 3D human pose estimation from monocular images require large amounts of in-the-wild 2D and controlled 3D pose annotated datasets which are costly and require sophisticated systems to acquire. To reduce this annotation dependency, we propose Multiview-Consistent Semi Supervised Learning (MCSS) framework that utilizes similarity in pose information from unannotated, uncalibrated but synchronized multi-view videos of human motions as additional weak supervision signal to guide 3D human pose regression. Our framework applies hard-negative mining based on temporal relations in multi-view videos to arrive at a multi-view consistent pose embedding. When jointly trained with limited 3D pose annotations, our approach improves the baseline by 25% and state-of-the-art by 8.7%, whilst using substantially smaller networks. Lastly, but importantly, we demonstrate the advantages of the learned embedding and establish view-invariant pose retrieval benchmarks on two popular, publicly available multi-view human pose datasets, Human 3.6M and MPI-INF-3DHP, to facilitate future research.

cs.CV↗

arXiv2020

Listwise Learning to Rank with Deep Q-Networks

Learning to Rank is the problem involved with ranking a sequence of documents based on their relevance to a given query. Deep Q-Learning has been shown to be a useful method for training an agent in sequential decision making. In this paper, we show that DeepQRank, our deep q-learning to rank agent, demonstrates performance that can be considered state-of-the-art. Though less computationally efficient than a supervised learning approach such as linear regression, our agent has fewer limitations in terms of which format of data it can use for training and evaluation. We run our algorithm against Microsoft's LETOR listwise dataset and achieve an NDCG@1 (ranking accuracy in the range [0,1]) of 0.5075, narrowly beating out the leading supervised learning model, SVMRank (0.4958).

cs.LG↗

arXiv2019

Analyzing Offline Social Engagements: An Empirical Study of Meetup Events Related to Software Development

Software developers use a variety of social media channels and tools in order to keep themselves up to date, collaborate with other developers, and find projects to contribute to. Meetup is one of such social media used by software developers to organize community gatherings. Liu et al. characterized Meetup as an event-based social network (EBSN) which contains valuable offline social interactions in addition to online interactions. Recently, Storey et al. found out that Meetup was one of the social channels used by developers. We in this work investigate in detail the dynamics of Meetup groups and events related to software development, which has not been done in any of the previous works. First, we identified 6,317 Meetup groups related to software development and extracted 185,758 events organized by them. Then we took a statistically significant sample of 452 events on which we performed open coding, based on which we were able to develop 9 categories of events (8 main categories + Others). Next, we did a popularity analysis of the categories of events and found that Talks by Domain Experts, Hands-on Sessions, and Open Discussions are the most popular categories of events organized by Meetup groups related to software development. Our findings show that more popular categories are those where developers can learn and gain knowledge. On doing a diversity analysis of Meetup groups we found 19.82% of the members on an average are female, which is a larger proportion as compared to numbers reported in previous studies on other social media. From a broader software development community point of view information from this new forum can be valuable to identify and understand emerging topics and associations among them which can be helpful to identify future trends as well as current best practices.

cs.SE↗

arXiv2019

Band-pass superlattice magnetic tunnel junctions

Significant scientific and technological progress in the field of spintronics is based on trilayer magnetic tunnel junction devices which principally rely on the physics of single barrier tunneling. While technologically relevant devices have been prototyped, the physics of single barrier tunneling poses ultimate limitations on the performance of magnetic tunnel junction devices. Here, we propose a fresh route toward high performance magnetic tunnel junctions by making electronic analogs of optical phenomena such as anti-reflections and Fabry-Pèrot resonances. The devices we propose feature anti-reflection enabled superlattice heterostructures sandwiched between the fixed and the free ferromagnets of the magnetic tunnel junction structure. Our predictions are based on the non-equilibrium Green's function spin transport formalism coupled self-consistently with the Landau-Lifshitz-Gilbert-Slonczewski equation. Owing to the physics of bandpass spin filtering in the bandpass superlattice magnetic tunnel junction device, we demonstrate an ultra-high boost in the tunnel magneto-resistance (TMR$\approx5\times10^4\%$) and nearly 92% suppression of spin transfer torque switching bias in comparison to a traditional trilayer magnetic tunnel junction device. We rationalize improvised spin transfer torque switching via analysis of the Slonczewski spin current transmission spectra. The proof of concepts presented here can lead to next-generation spintronics device design harvesting the rich physics of superlattice heterostructures and exploiting spintronic analogs of optical phenomena.

cond-mat.mes-hall↗

arXiv2019

Reinforcement learning with a network of spiking agents

Neuroscientific theory suggests that dopaminergic neurons broadcast global reward prediction errors to large areas of the brain influencing the synaptic plasticity of the neurons in those regions. We build on this theory to propose a multi-agent learning framework with spiking neurons in the generalized linear model (GLM) formulation as agents, to solve reinforcement learning (RL) tasks. We show that a network of GLM spiking agents connected in a hierarchical fashion, where each spiking agent modulates its firing policy based on local information and a global prediction error, can learn complex action representations to solve RL tasks. We further show how leveraging principles of modularity and population coding inspired from the brain can help reduce variance in the learning updates making it a viable optimization technique.

cs.LG↗

arXiv2019

Measurement of the relative response of TowerJazz Mini-MALTA CMOS prototypes at Diamond Light Source

This paper outlines the results of investigations into the effects of radiation damage in the mini-MALTA prototype. Measurements were carried out at Diamond Light Source using a micro-focus X-ray beam, which scanned across the surface of the device in 2 $\mathrm{μm}$ steps. This allowed the in-pixel photon response to be measured directly with high statistics. Three pixel design variations were considered: one with the standard continuous $\mathrm{n^-}$ layer layout and front-end, and extra deep p-well and $\mathrm{n^-}$ gap designs with a modified front-end. Five chips were measured: one unirradiated, one neutron irradiated, and three proton irradiated.

physics.ins-det↗

arXiv2019

Multi-Person 3D Human Pose Estimation from Monocular Images

Multi-person 3D human pose estimation from a single image is a challenging problem, especially for in-the-wild settings due to the lack of 3D annotated data. We propose HG-RCNN, a Mask-RCNN based network that also leverages the benefits of the Hourglass architecture for multi-person 3D Human Pose Estimation. A two-staged approach is presented that first estimates the 2D keypoints in every Region of Interest (RoI) and then lifts the estimated keypoints to 3D. Finally, the estimated 3D poses are placed in camera-coordinates using weak-perspective projection assumption and joint optimization of focal length and root translations. The result is a simple and modular network for multi-person 3D human pose estimation that does not require any multi-person 3D pose dataset. Despite its simple formulation, HG-RCNN achieves the state-of-the-art results on MuPoTS-3D while also approximating the 3D pose in the camera-coordinate system.

cs.CV↗

arXiv2019

ZoomOut: Spectral Upsampling for Efficient Shape Correspondence

We present a simple and efficient method for refining maps or correspondences by iterative upsampling in the spectral domain that can be implemented in a few lines of code. Our main observation is that high quality maps can be obtained even if the input correspondences are noisy or are encoded by a small number of coefficients in a spectral basis. We show how this approach can be used in conjunction with existing initialization techniques across a range of application scenarios, including symmetry detection, map refinement across complete shapes, non-rigid partial shape matching and function transfer. In each application we demonstrate an improvement with respect to both the quality of the results and the computational speed compared to the best competing methods, with up to two orders of magnitude speed-up in some applications. We also demonstrate that our method is both robust to noisy input and is scalable with respect to shape complexity. Finally, we present a theoretical justification for our approach, shedding light on structural properties of functional maps.

cs.GR↗