Streaming LifeLong Learning With Any-Time Inference

TL;DR

This paper introduces a novel streaming lifelong learning framework capable of single-pass, class-incremental learning with any-time inference, addressing dynamic environments and outperforming prior methods.

Contribution

It proposes a Bayesian-based streaming lifelong learning approach with implicit regularization, efficient sample selection, and replay buffer management for rapid adaptation and continual learning.

Findings

Outperforms prior methods by large margins

Enables fast parameter updates with single samples

Supports any-time inference in dynamic environments

Abstract

Despite rapid advancements in lifelong learning (LLL) research, a large body of research mainly focuses on improving the performance in the existing \textit{static} continual learning (CL) setups. These methods lack the ability to succeed in a rapidly changing \textit{dynamic} environment, where an AI agent needs to quickly learn new instances in a `single pass' from the non-i.i.d (also possibly temporally contiguous/coherent) data streams without suffering from catastrophic forgetting. For practical applicability, we propose a novel lifelong learning approach, which is streaming, i.e., a single input sample arrives in each time step, single pass, class-incremental, and subject to be evaluated at any moment. To address this challenging setup and various evaluation protocols, we propose a Bayesian framework, that enables fast parameter update, given a single training example, and enables any-time inference. We additionally propose an implicit regularizer in the form of snap-shot self-distillation, which effectively minimizes the forgetting further. We further propose an effective method that efficiently selects a subset of samples for online memory rehearsal and employs a new replay buffer management scheme that significantly boosts the overall performance. Our empirical evaluations and ablations demonstrate that the proposed method outperforms the prior works by large margins.