Drift to Remember

TL;DR

DriftNet introduces a novel lifelong learning approach inspired by biological neural representational drift, enabling AI models to continually learn new tasks while retaining prior knowledge efficiently, especially in large language models.

Contribution

This paper proposes DriftNet, a scalable neural network that leverages representational drift to improve lifelong learning in AI, particularly in large language models, without full retraining.

Findings

DriftNet outperforms existing models in image classification and NLP tasks.

It effectively handles sequences of tasks like sentiment analysis and question answering.

DriftNet is scalable and efficient for large models on single GPUs.

Abstract

Lifelong learning in artificial intelligence (AI) aims to mimic the biological brain's ability to continuously learn and retain knowledge, yet it faces challenges such as catastrophic forgetting. Recent neuroscience research suggests that neural activity in biological systems undergoes representational drift, where neural responses evolve over time, even with consistent inputs and tasks. We hypothesize that representational drift can alleviate catastrophic forgetting in AI during new task acquisition. To test this, we introduce DriftNet, a network designed to constantly explore various local minima in the loss landscape while dynamically retrieving relevant tasks. This approach ensures efficient integration of new information and preserves existing knowledge. Experimental studies in image classification and natural language processing demonstrate that DriftNet outperforms existing models in lifelong learning. Importantly, DriftNet is scalable in handling a sequence of tasks such as sentiment analysis and question answering using large language models (LLMs) with billions of parameters on a single Nvidia A100 GPU. DriftNet efficiently updates LLMs using only new data, avoiding the need for full dataset retraining. Tested on GPT-2 and RoBERTa, DriftNet is a robust, cost-effective solution for lifelong learning in LLMs. This study not only advances AI systems to emulate biological learning, but also provides insights into the adaptive mechanisms of biological neural systems, deepening our understanding of lifelong learning in nature.