arXiv subjects

Ayan Sengupta

Publications and source records attributed to Ayan Sengupta.

28 records · Page 2Linked to original sources

arXiv2021

HIT: A Hierarchically Fused Deep Attention Network for Robust Code-mixed Language Representation

Understanding linguistics and morphology of resource-scarce code-mixed texts remains a key challenge in text processing. Although word embedding comes in handy to support downstream tasks for low-resource languages, there are plenty of scopes in improving the quality of language representation particularly for code-mixed languages. In this paper, we propose HIT, a robust representation learning method for code-mixed texts. HIT is a hierarchical transformer-based framework that captures the semantic relationship among words and hierarchically learns the sentence-level semantics using a fused attention mechanism. HIT incorporates two attention modules, a multi-headed self-attention and an outer product attention module, and computes their weighted sum to obtain the attention weights. Our evaluation of HIT on one European (Spanish) and five Indic (Hindi, Bengali, Tamil, Telugu, and Malayalam) languages across four NLP tasks on eleven datasets suggests significant performance improvement against various state-of-the-art systems. We further show the adaptability of learned representation across tasks in a transfer learning setup (with and without fine-tuning).

cs.CL↗

arXiv2021

An Embedding-based Joint Sentiment-Topic Model for Short Texts

Short text is a popular avenue of sharing feedback, opinions and reviews on social media, e-commerce platforms, etc. Many companies need to extract meaningful information (which may include thematic content as well as semantic polarity) out of such short texts to understand users' behaviour. However, obtaining high quality sentiment-associated and human interpretable themes still remains a challenge for short texts. In this paper we develop ELJST, an embedding enhanced generative joint sentiment-topic model that can discover more coherent and diverse topics from short texts. It uses Markov Random Field Regularizer that can be seen as a generalisation of skip-gram based models. Further, it can leverage higher-order semantic information appearing in word embedding, such as self-attention weights in graphical models. Our results show an average improvement of 10% in topic coherence and 5% in topic diversification over baselines. Finally, ELJST helps understand users' behaviour at more granular levels which can be explained. All these can bring significant values to the service and healthcare industries often dealing with customers.

cs.CL↗

arXiv2021

An Autonomous Negotiating Agent Framework with Reinforcement Learning Based Strategies and Adaptive Strategy Switching Mechanism

Despite abundant negotiation strategies in literature, the complexity of automated negotiation forbids a single strategy from being dominant against all others in different negotiation scenarios. To overcome this, one approach is to use mixture of experts, but at the same time, one problem of this method is the selection of experts, as this approach is limited by the competency of the experts selected. Another problem with most negotiation strategies is their incapability of adapting to dynamic variation of the opponent's behaviour within a single negotiation session resulting in poor performance. This work focuses on both, solving the problem of expert selection and adapting to the opponent's behaviour with our Autonomous Negotiating Agent Framework. This framework allows real-time classification of opponent's behaviour and provides a mechanism to select, switch or combine strategies within a single negotiation session. Additionally, our framework has a reviewer component which enables self-enhancement capability by deciding to include new strategies or replace old ones with better strategies periodically. We demonstrate an instance of our framework by implementing maximum entropy reinforcement learning based strategies with a deep learning based opponent classifier. Finally, we evaluate the performance of our agent against state-of-the-art negotiators under varied negotiation scenarios.

cs.AI↗

arXiv2017

Wireless Network Simplification: The Performance of Routing

Consider a wireless Gaussian network where a source wishes to communicate with a destination with the help of N full-duplex relay nodes. Most practical systems today route information from the source to the destination using the best path that connects them. In this paper, we show that routing can in the worst case result in an unbounded gap from the network capacity - or reversely, physical layer cooperation can offer unbounded gains over routing. More specifically, we show that for $N$-relay Gaussian networks with an arbitrary topology, routing can in the worst case guarantee an approximate fraction $\frac{1}{\left\lfloor N/2 \right\rfloor + 1}$ of the capacity of the full network, independently of the SNR regime. We prove that this guarantee is fundamental, i.e., it is the highest worst-case guarantee that we can provide for routing in relay networks. Next, we consider how these guarantees are refined for Gaussian layered relay networks with $L$ layers and $N_L$ relays per layer. We prove that for arbitrary $L$ and $N_L$, there always exists a route in the network that approximately achieves at least $\frac{2}{(L-1)N_L + 4}$ $\left(\mbox{resp.}\frac{2}{LN_L+2}\right)$ of the network capacity for odd $L$ (resp. even $L$), and there exist networks where the best routes exactly achieve these fractions. These results are formulated within the network simplification framework, that asks what fraction of the capacity we can achieve by using a subnetwork (in our case, a single path). A fundamental step in our proof is a simplification result for MIMO antenna selection that may also be of independent interest. To the best of our knowledge, this is the first result that characterizes, for general wireless network topologies, what is the performance of routing with respect to physical layer cooperation techniques that approximately achieve the network capacity.

cs.IT↗

arXiv2016

Consistency in the face of change: an adaptive approach to physical layer cooperation

Most existing works on physical-layer (PHY) cooperation (beyond routing) focus on how to best use a given, static relay network--while wireless networks are anything but static. In this paper, we pose a different set of questions: given that we have multiple devices within range, which relay(s) do we use for PHY cooperation, to maintain a consistent target performance? How can we efficiently adapt, as network conditions change? And how important is it, in terms of performance, to adapt? Although adapting to the best path when routing is a well understood problem, how to do so over PHY cooperation networks is an open question. Our contributions are: (1) We demonstrate via theoretical evaluation, a diminishing returns trend as the number of deployed relays increases. (2) Using a simple algorithm based on network metrics, we efficiently select the sub-network to use at any given time to maintain a target reliability. (3) When streaming video from Netflix, we experimentally show (using measurements from a WARP radio testbed employing DIQIF relaying) that our adaptive PHY cooperation scheme provides a throughput gain of 2x over nonadaptive PHY schemes, and a gain of 6x over genie-aided IP-level adaptive routing.

cs.NI↗

arXiv2016

Fault Detection Engine in Intelligent Predictive Analytics Platform for DCIM

With the advancement of huge data generation and data handling capability, Machine Learning and Probabilistic modelling enables an immense opportunity to employ predictive analytics platform in high security critical industries namely data centers, electricity grids, utilities, airport etc. where downtime minimization is one of the primary objectives. This paper proposes a novel, complete architecture of an intelligent predictive analytics platform, Fault Engine, for huge device network connected with electrical/information flow. Three unique modules, here proposed, seamlessly integrate with available technology stack of data handling and connect with middleware to produce online intelligent prediction in critical failure scenarios. The Markov Failure module predicts the severity of a failure along with survival probability of a device at any given instances. The Root Cause Analysis model indicates probable devices as potential root cause employing Bayesian probability assignment and topological sort. Finally, a community detection algorithm produces correlated clusters of device in terms of failure probability which will further narrow down the search space of finding route cause. The whole Engine has been tested with different size of network with simulated failure environments and shows its potential to be scalable in real-time implementation.

cs.AI↗

arXiv2016

A Large-scale Analysis of the Marketplace Characteristics in Fiverr

Crowdsourcing platforms have become quite popular due to the increasing demand of human computation-based tasks. Though the crowdsourcing systems are primarily demand-driven like MTurk, supply-driven marketplaces are becoming increasingly popular. Fiverr is a fast growing supply-driven marketplace where the sellers post micro-tasks (gigs) and users purchase them for prices as low as $5. In this paper, we study the Fiverr platform as a unique marketplace and characterize the sellers, buyers and the interactions among them. We find that sellers are more appeasing in their interactions and try to woo their buyers into buying their gigs. There are many small tightly-knit communities existing in the seller-seller network who support each other. We also study Fiverr as a seller-driven marketplace in terms of sales, churn rates, competitiveness among various sub-categories etc. and observe that while there are certain similarities with common marketplaces there are also many differences.

cs.SI↗

arXiv2016

A Note on Antenna Selection in Gaussian MIMO Channels: Capacity Guarantees and Bounds

We consider the problem of selecting $k_t \times k_r$ antennas from a Gaussian MIMO channel with $n_t \times n_r$ antennas, where $k_t \leq n_t$ and $k_r \leq n_r$. We prove the following two results that hold universally, in the sense that they do not depend on the channel coefficients: (i) The capacity of the best $k_t \times k_r$ subchannel is always lower bounded by a fraction $\frac{k_t k_r}{n_t n_r}$ of the full capacity (with $n_t \times n_r$ antennas). This bound is tight as the channel coefficients diminish in magnitude. (ii) There always exists a selection of $k_t \times k_r$ antennas (including the best) that achieves a fraction greater than $\frac{\min(k_t ,k_r)}{\min(n_t,n_r)}$ of the full capacity within an additive constant that is independent of the coefficients in the channel matrix. The key mathematical idea that allows us to derive these universal bounds is to directly relate the determinants of principle sub-matrices of a Hermitian matrix to the determinant of the entire matrix.

cs.IT↗

arXiv2016

Wireless Network Simplification : Beyond Diamond Networks

We consider an arbitrary layered Gaussian relay network with $L$ layers of $N$ relays each, from which we select subnetworks with $K$ relays per layer. We prove that: (i) For arbitrary $L, N$ and $K = 1$, there always exists a subnetwork that approximately achieves $\frac{2}{(L-1)N + 4}$ $\left(\mbox{resp.}\frac{2}{LN+2}\right)$ of the network capacity for odd $L$ (resp. even $L$), (ii) For $L = 2, N = 3, K = 2$, there always exists a subnetwork that approximately achieves $\frac{1}{2}$ of the network capacity. We also provide example networks where even the best subnetworks achieve exactly these fractions (up to additive gaps). Along the way, we derive some results on MIMO antenna selection and capacity decomposition that may also be of independent interest.

cs.IT↗

arXiv2013

Cooperative Relaying at Finite SNR -- Role of Quantize-Map-and-Forward

Quantize-Map-and-Forward (QMF) relaying has been shown to achieve the optimal diversity-multiplexing trade-off (DMT) for arbitrary slow fading full-duplex networks as well as for the single-relay half-duplex network. A key reason for this is that quantizing at the noise level suffices to achieve the cut-set bound approximately to within an additive gap, without any requirement of instantaneous channel state information (CSI). However, DMT only captures the high SNR performance and potentially, limited CSI at the relay can improve performance at moderate SNRs. In this work we propose an optimization framework for QMF relaying over slow fading channels. Focusing on vector Gaussian quantizers, we optimize the outage probability for the full-duplex and half-duplex single relay by finding the best quantization level and relay schedule according to the available CSI at the relays. For the N-relay diamond network, we derive an universal quantizer that sharpens the additive approximation gap of QMF from the conventional Θ(N) bits/s/Hz to Θ(log(N)) bits/s/Hz using only network topology information. Analytical solutions to channel-aware optimal quantizers for two-relay and symmetric N-relay diamond networks are also derived. In addition, we prove that suitable hybridizations of our optimized QMF schemes with Decode-Forward (DF) or Dynamic DF protocols provide significant finite SNR gains over the individual schemes.

cs.IT↗