Výsledky vyhledávání - "Verma, Vinay K"

Report

Resource Efficient Perception for Vision Systems

Autor: Subramanyam, A V, Singal, Niyati, Verma, Vinay K

Despite the rapid advancement in the field of image recognition, the processing of high-resolution imagery remains a computational challenge. However, this processing is pivotal for extracting detailed object insights in areas ranging from autonomous

Externí odkaz: http://arxiv.org/abs/2405.07166

Zobrazit plný text záznamu

Report

Convolutional Prompting meets Language Models for Continual Learning

Autor: Roy, Anurag, Moulick, Riddhiman, Verma, Vinay K., Ghosh, Saptarshi, Das, Abir

Continual Learning (CL) enables machine learning models to learn from continuously shifting new training data in absence of data from old tasks. Recently, pretrained vision transformers combined with prompt tuning have shown promise for overcoming ca

Externí odkaz: http://arxiv.org/abs/2403.20317

Zobrazit plný text záznamu

Report

Efficient Expansion and Gradient Based Task Inference for Replay Free Incremental Learning

Autor: Roy, Soumya, Verma, Vinay K, Gupta, Deepak

This paper proposes a simple but highly efficient expansion-based model for continual learning. The recent feature transformation, masking and factorization-based methods are efficient, but they grow the model only over the global or shared parameter

Externí odkaz: http://arxiv.org/abs/2312.01188

Zobrazit plný text záznamu

Report

Meta-Learned Attribute Self-Interaction Network for Continual and Generalized Zero-Shot Learning

Autor: Verma, Vinay K, Mehta, Nikhil, Liang, Kevin J, Mishra, Aakansha, Carin, Lawrence

Zero-shot learning (ZSL) is a promising approach to generalizing a model to categories unseen during training by leveraging class attributes, but challenges remain. Recently, methods using generative models to combat bias towards classes seen during

Externí odkaz: http://arxiv.org/abs/2312.01167

Zobrazit plný text záznamu

Report

VERSE: Virtual-Gradient Aware Streaming Lifelong Learning with Anytime Inference

Autor: Banerjee, Soumya, Verma, Vinay K., Mukherjee, Avideep, Gupta, Deepak, Namboodiri, Vinay P., Rai, Piyush

Lifelong learning or continual learning is the problem of training an AI agent continuously while also preventing it from forgetting its previously acquired knowledge. Streaming lifelong learning is a challenging setting of lifelong learning with the

Externí odkaz: http://arxiv.org/abs/2309.08227

Zobrazit plný text záznamu

Report

Continual Learning using a Bayesian Nonparametric Dictionary of Weight Factors

Autor: Mehta, Nikhil, Liang, Kevin J, Verma, Vinay K, Carin, Lawrence

Naively trained neural networks tend to experience catastrophic forgetting in sequential task settings, where data from previous tasks are unavailable. A number of methods, using various model expansion strategies, have been proposed recently as poss

Externí odkaz: http://arxiv.org/abs/2004.10098

Zobrazit plný text záznamu