Saved in:
Bibliographic Details
Main Authors: Banerjee, Soumya, Verma, Vinay K., Mukherjee, Avideep, Gupta, Deepak, Namboodiri, Vinay P., Rai, Piyush
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2309.08227
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917592486117376
author Banerjee, Soumya
Verma, Vinay K.
Mukherjee, Avideep
Gupta, Deepak
Namboodiri, Vinay P.
Rai, Piyush
author_facet Banerjee, Soumya
Verma, Vinay K.
Mukherjee, Avideep
Gupta, Deepak
Namboodiri, Vinay P.
Rai, Piyush
contents 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 goal of continuous learning in a dynamic non-stationary environment without forgetting. We introduce a novel approach to lifelong learning, which is streaming (observes each training example only once), requires a single pass over the data, can learn in a class-incremental manner, and can be evaluated on-the-fly (anytime inference). To accomplish these, we propose a novel \emph{virtual gradients} based approach for continual representation learning which adapts to each new example while also generalizing well on past data to prevent catastrophic forgetting. Our approach also leverages an exponential-moving-average-based semantic memory to further enhance performance. Experiments on diverse datasets with temporally correlated observations demonstrate our method's efficacy and superior performance over existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2309_08227
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle VERSE: Virtual-Gradient Aware Streaming Lifelong Learning with Anytime Inference
Banerjee, Soumya
Verma, Vinay K.
Mukherjee, Avideep
Gupta, Deepak
Namboodiri, Vinay P.
Rai, Piyush
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
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 goal of continuous learning in a dynamic non-stationary environment without forgetting. We introduce a novel approach to lifelong learning, which is streaming (observes each training example only once), requires a single pass over the data, can learn in a class-incremental manner, and can be evaluated on-the-fly (anytime inference). To accomplish these, we propose a novel \emph{virtual gradients} based approach for continual representation learning which adapts to each new example while also generalizing well on past data to prevent catastrophic forgetting. Our approach also leverages an exponential-moving-average-based semantic memory to further enhance performance. Experiments on diverse datasets with temporally correlated observations demonstrate our method's efficacy and superior performance over existing methods.
title VERSE: Virtual-Gradient Aware Streaming Lifelong Learning with Anytime Inference
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2309.08227