Batch Training vs. Online Learning

cover
27 Aug 2024

Authors:

(1) Sebastian Dziadzio, University of Tübingen (sebastian.dziadzio@uni-tuebingen.de);

(2) Çagatay Yıldız, University of Tübingen;

(3) Gido M. van de Ven, KU Leuven;

(4) Tomasz Trzcinski, IDEAS NCBR, Warsaw University of Technology, Tooploox;

(5) Tinne Tuytelaars, KU Leuven;

(6) Matthias Bethge, University of Tübingen.

Abstract and 1. Introduction

2. Two problems with the current approach to class-incremental continual learning

3. Methods and 3.1. Infinite dSprites

3.2. Disentangled learning

4. Related work

4.1. Continual learning and 4.2. Benchmarking continual learning

5. Experiments

5.1. Regularization methods and 5.2. Replay-based methods

5.3. Do we need equivariance?

5.4. One-shot generalization and 5.5. Open-set classification

5.6. Online vs. offline

Conclusion, Acknowledgments and References

Supplementary Material

5.6. Online vs. offline

In all previous experiments, we applied our method in batch mode: we performed multiple training passes over the data for each task. However, efficiently learning from streaming data might require observing each training sample only once to make sure computation is not becoming a bottleneck. This is why we test our method in the online learning regime and compare it to two batch learning scenarios. The results are shown in Fig. 9. Unsurprisingly, training for multiple epochs results in better and more robust accuracy on past tasks; it is however worth noting that our method still improves over time in the online learning scenario. It is possible that, given enough tasks, all three curves would converge.

Figure 8. Some of the incorrectly classified unseen shapes in the open-set recognition task. Top: outputs of the normalization module. Bottom: closest training exemplars.

Figure 9. Cumulative test accuracy of our method in the online scenario (every data point is only seen once) and two offline scenarios (3 and 5 training epochs per task). Each plot is an average of five runs.

This paper is available on arxiv under CC 4.0 license.