arXiv subjects

Ayush Maheshwari

Publications and source records attributed to Ayush Maheshwari.

23 records · Page 2Linked to original sources

arXiv2021

Rule Augmented Unsupervised Constituency Parsing

Recently, unsupervised parsing of syntactic trees has gained considerable attention. A prototypical approach to such unsupervised parsing employs reinforcement learning and auto-encoders. However, no mechanism ensures that the learnt model leverages the well-understood language grammar. We propose an approach that utilizes very generic linguistic knowledge of the language present in the form of syntactic rules, thus inducing better syntactic structures. We introduce a novel formulation that takes advantage of the syntactic grammar rules and is independent of the base system. We achieve new state-of-the-art results on two benchmarks datasets, MNLI and WSJ. The source code of the paper is available at https://github.com/anshuln/Diora_with_rules.

cs.CL↗

arXiv2021

Unsupervised Learning of Explainable Parse Trees for Improved Generalisation

Recursive neural networks (RvNN) have been shown useful for learning sentence representations and helped achieve competitive performance on several natural language inference tasks. However, recent RvNN-based models fail to learn simple grammar and meaningful semantics in their intermediate tree representation. In this work, we propose an attention mechanism over Tree-LSTMs to learn more meaningful and explainable parse tree structures. We also demonstrate the superior performance of our proposed model on natural language inference, semantic relatedness, and sentiment analysis tasks and compare them with other state-of-the-art RvNN based methods. Further, we present a detailed qualitative and quantitative analysis of the learned parse trees and show that the discovered linguistic structures are more explainable, semantically meaningful, and grammatically correct than recent approaches. The source code of the paper is available at https://github.com/atul04/Explainable-Latent-Structures-Using-Attention.

cs.CL↗

arXiv2021

Joint Learning of Hyperbolic Label Embeddings for Hierarchical Multi-label Classification

We consider the problem of multi-label classification where the labels lie in a hierarchy. However, unlike most existing works in hierarchical multi-label classification, we do not assume that the label-hierarchy is known. Encouraged by the recent success of hyperbolic embeddings in capturing hierarchical relations, we propose to jointly learn the classifier parameters as well as the label embeddings. Such a joint learning is expected to provide a twofold advantage: i) the classifier generalizes better as it leverages the prior knowledge of existence of a hierarchy over the labels, and ii) in addition to the label co-occurrence information, the label-embedding may benefit from the manifold structure of the input datapoints, leading to embeddings that are more faithful to the label hierarchy. We propose a novel formulation for the joint learning and empirically evaluate its efficacy. The results show that the joint learning improves over the baseline that employs label co-occurrence based pre-trained hyperbolic embeddings. Moreover, the proposed classifiers achieve state-of-the-art generalization on standard benchmarks. We also present evaluation of the hyperbolic embeddings obtained by joint learning and show that they represent the hierarchy more accurately than the other alternatives.

cs.LG↗

arXiv2020

Tale of tails using rule augmented sequence labeling for event extraction

The problem of event extraction is a relatively difficult task for low resource languages due to the non-availability of sufficient annotated data. Moreover, the task becomes complex for tail (rarely occurring) labels wherein extremely less data is available. In this paper, we present a new dataset (InDEE-2019) in the disaster domain for multiple Indic languages, collected from news websites. Using this dataset, we evaluate several rule-based mechanisms to augment deep learning based models. We formulate our problem of event extraction as a sequence labeling task and perform extensive experiments to study and understand the effectiveness of different approaches. We further show that tail labels can be easily incorporated by creating new rules without the requirement of large annotated data.

cs.IR↗

arXiv2019

Peak Electricity Demand and Global Warming in the Industrial and Residential areas of Pune : An Extreme Value Approach

Industrial and residential activities respond distinctly to electricity demand on temperature. Due to increasing temperature trend on account of global warming, its impact on peak electricity demand is a proxy for effective management of electricity infrastructure. Few studies explore the relationship between electricity demand and temperature changes in industrial areas in India mainly due to the limitation of data. The precise role of industrial and residential activities response to the temperature is not explored in sub-tropical humid climate of India. Here, we show the temperature sensitivity of industrial and residential areas in the city of Pune, Maharashtra by keeping other influencing variables on electricity demand as constant. The study seeks to estimate the behaviour of peak electricity demand with the apparent temperature (AT) using the Extreme Value Theory. Our analysis shows that industrial activities are not much influenced by the temperature whereas residential activities show around 1.5-2% change in average electricity demand with 1 degree rise in AT. Further, we show that peak electricity demand in residential areas, performed using stationary and non-stationary GEV models, are significantly influenced by the rise in temperature. The study shows that with the improvement in data collection, better planning for the future development, accounting for the climate change effects, will enhance the effectiveness of electricity distribution system. The study is limited to the geographical area of Pune. However, the methods are useful in estimating the peak power load attributed to climate change to other geographical regions located in subtropical and humid climate.

stat.AP↗