arXiv subjects

Shubham Gupta

Publications and source records attributed to Shubham Gupta.

79 records · Page 5Linked to original sources

arXiv2020

A Large-Scale Deep Architecture for Personalized Grocery Basket Recommendations

With growing consumer adoption of online grocery shopping through platforms such as Amazon Fresh, Instacart, and Walmart Grocery, there is a pressing business need to provide relevant recommendations throughout the customer journey. In this paper, we introduce a production within-basket grocery recommendation system, RTT2Vec, which generates real-time personalized product recommendations to supplement the user's current grocery basket. We conduct extensive offline evaluation of our system and demonstrate a 9.4% uplift in prediction metrics over baseline state-of-the-art within-basket recommendation models. We also propose an approximate inference technique 11.6x times faster than exact inference approaches. In production, our system has resulted in an increase in average basket size, improved product discovery, and enabled faster user check-out

cs.IR↗

arXiv2018

A Generative Model for Dynamic Networks with Applications

Networks observed in real world like social networks, collaboration networks etc., exhibit temporal dynamics, i.e. nodes and edges appear and/or disappear over time. In this paper, we propose a generative, latent space based, statistical model for such networks (called dynamic networks). We consider the case where the number of nodes is fixed, but the presence of edges can vary over time. Our model allows the number of communities in the network to be different at different time steps. We use a neural network based methodology to perform approximate inference in the proposed model and its simplified version. Experiments done on synthetic and real world networks for the task of community detection and link prediction demonstrate the utility and effectiveness of our model as compared to other similar existing approaches.

cs.SI↗

arXiv2018

InfyNLP at SMM4H Task 2: Stacked Ensemble of Shallow Convolutional Neural Networks for Identifying Personal Medication Intake from Twitter

This paper describes Infosys's participation in the "2nd Social Media Mining for Health Applications Shared Task at AMIA, 2017, Task 2". Mining social media messages for health and drug related information has received significant interest in pharmacovigilance research. This task targets at developing automated classification models for identifying tweets containing descriptions of personal intake of medicines. Towards this objective we train a stacked ensemble of shallow convolutional neural network (CNN) models on an annotated dataset provided by the organizers. We use random search for tuning the hyper-parameters of the CNN and submit an ensemble of best models for the prediction task. Our system secured first place among 9 teams, with a micro-averaged F-score of 0.693.

cs.CL↗

arXiv2017

Attentive Recurrent Comparators

Rapid learning requires flexible representations to quickly adopt to new evidence. We develop a novel class of models called Attentive Recurrent Comparators (ARCs) that form representations of objects by cycling through them and making observations. Using the representations extracted by ARCs, we develop a way of approximating a \textit{dynamic representation space} and use it for one-shot learning. In the task of one-shot classification on the Omniglot dataset, we achieve the state of the art performance with an error rate of 1.5\%. This represents the first super-human result achieved for this task with a generic model that uses only pixel information.

cs.CV↗

arXiv2015

Synthesis of Sequential Reversible Circuits through Finite State Machine

Reversible computing has attracted the attention of researchers due to its low power consumption and less heat dissipation compared to conventional computing. A number of reversible gates have been proposed by different researchers and various combinational circuits based on reversible gates have been developed. However the realization of sequential circuit in reversible logic is still at premature stage. Sequential circuits were not available because of feedback was not allowed in reversible circuit. However allowing feedback in space (not in time), some sequential reversible gates and circuits have been reported in the literature. In this dissertation, we have addressed the problem from two sides. One side is to propose a low cost reversible gate suitable for sequential building block i.e. T flip-flop and hence designing low cost synchronous and asynchronous counters. Another side is to generate the circuit from its behavioral description described in FSM form. Our propose designs of reversible counters are significantly better in optimization parameters such as gate counts, garbage outputs and constant inputs available in literature. We have also proposed a procedure for obtaining reversible circuit from behavioral description through FSM. A very few attempts have been reported in the literature for the conversion FSM to reversible FSM.Because of non-availability of generated sequential reversible circuit in literature, our results cannot be compared with any other circuits. We expect that the sequential reversible circuits will help in debugging the reversible circuits, handling the ambiguous state of an FSM and generating the original input in reverse direction by reversing the original output.

cs.ET↗

arXiv2014

Approximation algorithms for Capacitated Facility Location Problem with Penalties

In this paper, we address the problem of capacitated facility location problem with penalties (CapFLPP) paid per unit of unserved demand. In case of uncapacitated FLP with penalties demands of a client are either entirely met or are entirely rejected and penalty is paid. In the uncapacitated case, there is no reason to serve a client partially. Whereas, in case of CapFLPP, it may be beneficial to serve a client partially instead of not serving at all and, pay the penalty for the unmet demand. Charikar et. al. \cite{charikar2001algorithms}, Jain et. al. \cite{jain2003greedy} and Xu- Xu \cite{xu2009improved} gave $3$, $2$ and $1.8526$ approximation, respectively, for the uncapacitated case . We present $(5.83 + ε)$ factor for the case of uniform capacities and $(8.532 + ε)$ factor for non-uniform capacities.

cs.DS↗

arXiv2013

CPU and/or GPU: Revisiting the GPU Vs. CPU Myth

Parallel computing using accelerators has gained widespread research attention in the past few years. In particular, using GPUs for general purpose computing has brought forth several success stories with respect to time taken, cost, power, and other metrics. However, accelerator based computing has signifi- cantly relegated the role of CPUs in computation. As CPUs evolve and also offer matching computational resources, it is important to also include CPUs in the computation. We call this the hybrid computing model. Indeed, most computer systems of the present age offer a degree of heterogeneity and therefore such a model is quite natural. We reevaluate the claim of a recent paper by Lee et al.(ISCA 2010). We argue that the right question arising out of Lee et al. (ISCA 2010) should be how to use a CPU+GPU platform efficiently, instead of whether one should use a CPU or a GPU exclusively. To this end, we experiment with a set of 13 diverse workloads ranging from databases, image processing, sparse matrix kernels, and graphs. We experiment with two different hybrid platforms: one consisting of a 6-core Intel i7-980X CPU and an NVidia Tesla T10 GPU, and another consisting of an Intel E7400 dual core CPU with an NVidia GT520 GPU. On both these platforms, we show that hybrid solutions offer good advantage over CPU or GPU alone solutions. On both these platforms, we also show that our solutions are 90% resource efficient on average. Our work therefore suggests that hybrid computing can offer tremendous advantages at not only research-scale platforms but also the more realistic scale systems with significant performance gains and resource efficiency to the large scale user community.

cs.DC↗