Markus Weißflog
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Markus Weißflog has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Peer Neubert
University of Koblenz
2 shared publications - Peter Protzel
Chemnitz University of Technology
2 shared publications Peer Neubert
1 shared publicationPeter Protzel
Faculty of Electrical Engineering and Information Technology Chemnitz University of Technology Germany
1 shared publication- Stefan Schubert
TU Chemnitz
1 shared publication
Latest Publications
- 30 December 2024
- Published by Springer Nature
- p. 94-106
Towards Revisiting Visual Place Recognition for Joining Submaps in Multimap SLAM$^\star $- 30 December 2024
- Published by Springer Nature
- p. 94-106
Towards Revisiting Visual Place Recognition for Joining Submaps in Multimap SLAM$^\star $- Open Access
FETCH: A Memory-Efficient Replay Approach for Continual Learning in Image Classification
Class-incremental continual learning is an important area of research, as static deep learning methods fail to adapt to changing tasks and data distributions. In previous works, promising results were achieved using replay and compressed replay techniques. In the field of regular replay, GDumb achieved outstanding results but requires a large amount of memory. This problem can be addressed by compressed replay techniques. The goal of this work is to evaluate compressed replay in the pipeline of GDumb. We propose FETCH, a two-stage compression approach. First, the samples from the continual datastream are encoded by the early layers of a pre-trained neural network. Second, the samples are compressed before being stored in the episodic memory. Following GDumb, the remaining classification head is trained from scratch using only the decompressed samples from the reply memory. We evaluate FETCH in different scenarios and show that this approach can increase accuracy on CIFAR10 and CIFAR100. In our experiments, simple compression methods (e.g., quantization of tensors) outperform deep autoencoders. In the future, FETCH could serve as a baseline for benchmarking compressed replay learning in constrained memory scenarios.
Open AccessFETCH: A Memory-Efficient Replay Approach for Continual Learning in Image Classification
Class-incremental continual learning is an important area of research, as static deep learning methods fail to adapt to changing tasks and data distributions. In previous works, promising results were achieved using replay and compressed replay techniques. In the field of regular replay, GDumb achieved outstanding results but requires a large amount of memory. This problem can be addressed by compressed replay techniques. The goal of this work is to evaluate compressed replay in the pipeline of GDumb. We propose FETCH, a two-stage compression approach. First, the samples from the continual datastream are encoded by the early layers of a pre-trained neural network. Second, the samples are compressed before being stored in the episodic memory. Following GDumb, the remaining classification head is trained from scratch using only the decompressed samples from the reply memory. We evaluate FETCH in different scenarios and show that this approach can increase accuracy on CIFAR10 and CIFAR100. In our experiments, simple compression methods (e.g., quantization of tensors) outperform deep autoencoders. In the future, FETCH could serve as a baseline for benchmarking compressed replay learning in constrained memory scenarios.
- 15 November 2023
- Published by Springer Nature
- p. 418-430
FETCH: A Memory-Efficient Replay Approach for Continual Learning in Image Classification
- 15 November 2023
- Published by Springer Nature
- p. 418-430
FETCH: A Memory-Efficient Replay Approach for Continual Learning in Image Classification