Arnav Varma
Arnav Varma has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Elahe Arani
Eindhoven University of Technology
2 shared publications Adrian Staniec
Navinfo Europe
1 shared publication- Ahmed Badar
Navinfo Europe
1 shared publication Bahram Zonooz
Navinfo Europe
1 shared publication- Bahram Zonooz
Advanced Research Lab NavInfo Europe Eindhoven The Netherlands
1 shared publication - Bahram Zonooz
Advanced Research Lab, NavInfo Europe
1 shared publication
Latest Publications
- 18 October 2023
- Published by Springer Nature
- p. 281-303
Transformers in Unsupervised Structure-from-Motion
- 18 October 2023
- Published by Springer Nature
- p. 281-303
Transformers in Unsupervised Structure-from-Motion
- Open Access
Dynamically Modular and Sparse General Continual Learning
Real-world applications often require learning continuously from a stream of data under ever-changing conditions. When trying to learn from such non-stationary data, deep neural networks (DNNs) undergo catastrophic forgetting of previously learned information. Among the common approaches to avoid catastrophic forgetting, rehearsal-based methods have proven effective. However, they are still prone to forgetting due to task-interference as all parameters respond to all tasks. To counter this, we take inspiration from sparse coding in the brain and introduce dynamic modularity and sparsity (Dynamos) for rehearsal-based general continual learning. In this setup, the DNN learns to respond to stimuli by activating relevant subsets of neurons. We demonstrate the effectiveness of Dynamos on multiple datasets under challenging continual learning evaluation protocols. Finally, we show that our method learns representations that are modular and specialized, while maintaining reusability by activating subsets of neurons with overlaps corresponding to the similarity of stimuli.
Open AccessDynamically Modular and Sparse General Continual Learning
Real-world applications often require learning continuously from a stream of data under ever-changing conditions. When trying to learn from such non-stationary data, deep neural networks (DNNs) undergo catastrophic forgetting of previously learned information. Among the common approaches to avoid catastrophic forgetting, rehearsal-based methods have proven effective. However, they are still prone to forgetting due to task-interference as all parameters respond to all tasks. To counter this, we take inspiration from sparse coding in the brain and introduce dynamic modularity and sparsity (Dynamos) for rehearsal-based general continual learning. In this setup, the DNN learns to respond to stimuli by activating relevant subsets of neurons. We demonstrate the effectiveness of Dynamos on multiple datasets under challenging continual learning evaluation protocols. Finally, we show that our method learns representations that are modular and specialized, while maintaining reusability by activating subsets of neurons with overlaps corresponding to the similarity of stimuli.
- 19 July 2022
- Published by Springer Nature
- p. 515-531
Highlighting the Importance of Reducing Research Bias and Carbon Emissions in CNNs
- 19 July 2022
- Published by Springer Nature
- p. 515-531
Highlighting the Importance of Reducing Research Bias and Carbon Emissions in CNNs