CL2R: Compatible Lifelong Learning Representations

TL;DR

This paper introduces CL2R, a method for lifelong learning that maintains compatible feature representations over time, enabling effective visual search without forgetting previous knowledge.

Contribution

The paper proposes a novel training procedure promoting stationarity in representations, ensuring compatibility and robustness in lifelong learning scenarios.

Findings

Outperforms existing baselines and state-of-the-art methods on benchmark datasets.

Introduces new metrics for evaluating compatible representations under catastrophic forgetting.

Demonstrates the effectiveness of stationarity in maintaining feature compatibility.

Abstract

In this paper, we propose a method to partially mimic natural intelligence for the problem of lifelong learning representations that are compatible. We take the perspective of a learning agent that is interested in recognizing object instances in an open dynamic universe in a way in which any update to its internal feature representation does not render the features in the gallery unusable for visual search. We refer to this learning problem as Compatible Lifelong Learning Representations (CL2R) as it considers compatible representation learning within the lifelong learning paradigm. We identify stationarity as the property that the feature representation is required to hold to achieve compatibility and propose a novel training procedure that encourages local and global stationarity on the learned representation. Due to stationarity, the statistical properties of the learned features do not change over time, making them interoperable with previously learned features. Extensive experiments on standard benchmark datasets show that our CL2R training procedure outperforms alternative baselines and state-of-the-art methods. We also provide novel metrics to specifically evaluate compatible representation learning under catastrophic forgetting in various sequential learning tasks. Code at https://github.com/NiccoBiondi/CompatibleLifelongRepresentation.