Výsledky vyhledávání - "SURESH, VARSHA"

Report

An Adapter-Based Unified Model for Multiple Spoken Language Processing Tasks

Autor: Suresh, Varsha, Aït-Mokhtar, Salah, Brun, Caroline, Calapodescu, Ioan

Self-supervised learning models have revolutionized the field of speech processing. However, the process of fine-tuning these models on downstream tasks requires substantial computational resources, particularly when dealing with multiple speech-proc

Externí odkaz: http://arxiv.org/abs/2406.14747

Zobrazit plný text záznamu

Report

Using Positive Matching Contrastive Loss with Facial Action Units to mitigate bias in Facial Expression Recognition

Autor: Suresh, Varsha, Ong, Desmond C.

Publikováno v: 10th International Conference on Affective Computing and Intelligent Interaction (ACII), 2022

Machine learning models automatically learn discriminative features from the data, and are therefore susceptible to learn strongly-correlated biases, such as using protected attributes like gender and race. Most existing bias mitigation approaches ai

Externí odkaz: http://arxiv.org/abs/2303.04896

Zobrazit plný text záznamu

Report

Not All Negatives are Equal: Label-Aware Contrastive Loss for Fine-grained Text Classification

Autor: Suresh, Varsha, Ong, Desmond C.

Fine-grained classification involves dealing with datasets with larger number of classes with subtle differences between them. Guiding the model to focus on differentiating dimensions between these commonly confusable classes is key to improving perf

Externí odkaz: http://arxiv.org/abs/2109.05427

Zobrazit plný text záznamu

Report

Using Knowledge-Embedded Attention to Augment Pre-trained Language Models for Fine-Grained Emotion Recognition

Autor: Suresh, Varsha, Ong, Desmond C.

Modern emotion recognition systems are trained to recognize only a small set of emotions, and hence fail to capture the broad spectrum of emotions people experience and express in daily life. In order to engage in more empathetic interactions, future

Externí odkaz: http://arxiv.org/abs/2108.00194

Zobrazit plný text záznamu

Report

Critically examining the Domain Generalizability of Facial Expression Recognition models

Autor: Suresh, Varsha, Yeo, Gerard, Ong, Desmond C.

Facial Expression Recognition is a commercially-important application, but one under-appreciated limitation is that such applications require making predictions on out-of-sample distributions, where target images have different properties from the im

Externí odkaz: http://arxiv.org/abs/2106.15453

Zobrazit plný text záznamu

Report

Shape-CD: Change-Point Detection in Time-Series Data with Shapes and Neurons

Autor: Suresh, Varsha, Ooi, Wei Tsang

Change-point detection in a time series aims to discover the time points at which some unknown underlying physical process that generates the time-series data has changed. We found that existing approaches become less accurate when the underlying pro

Externí odkaz: http://arxiv.org/abs/2007.11985

Zobrazit plný text záznamu