Sample Weight Estimation Using Meta-Updates for Online Continual Learning

TL;DR

This paper proposes OMSI, a meta-learning strategy to estimate sample weights dynamically in online continual learning, improving performance especially in noisy and complex data scenarios.

Contribution

It introduces OMSI, a novel meta-learning approach for online sample weight estimation in continual learning, addressing limitations of uniform weighting strategies.

Findings

OMSI improves accuracy in noisy data streams.

OMSI outperforms existing replay strategies on benchmarks.

Sample weighting enhances continual learning robustness.

Abstract

The loss function plays an important role in optimizing the performance of a learning system. A crucial aspect of the loss function is the assignment of sample weights within a mini-batch during loss computation. In the context of continual learning (CL), most existing strategies uniformly treat samples when calculating the loss value, thereby assigning equal weights to each sample. While this approach can be effective in certain standard benchmarks, its optimal effectiveness, particularly in more complex scenarios, remains underexplored. This is particularly pertinent in training "in the wild," such as with self-training, where labeling is automated using a reference model. This paper introduces the Online Meta-learning for Sample Importance (OMSI) strategy that approximates sample weights for a mini-batch in an online CL stream using an inner- and meta-update mechanism. This is done by first estimating sample weight parameters for each sample in the mini-batch, then, updating the model with the adapted sample weights. We evaluate OMSI in two distinct experimental settings. First, we show that OMSI enhances both learning and retained accuracy in a controlled noisy-labeled data stream. Then, we test the strategy in three standard benchmarks and compare it with other popular replay-based strategies. This research aims to foster the ongoing exploration in the area of self-adaptive CL.