SCALE
Scalable Adversarial Online Continual Learning
Abstract: dversarial continual learning is effective for continual learning prob- lems because of the presence of feature alignment process generating task-invariant features having low susceptibility to the catastrophic forgetting problem. Never- theless, the ACL method imposes considerable complexities because it relies on task-specific networks and discriminators. It also goes through an iterative train- ing process which does not fit for online (one-epoch) continual learning problems. This paper proposes a scalable adversarial continual learning (SCALE) method putting forward a parameter generator transforming common features into task- specific features and a single discriminator in the adversarial game to induce common features. The training process is carried out in meta-learning fashions using a new combination of three loss functions. SCALE outperforms prominent baselines with noticeable margins in both accuracy and execution time.
Authors: Tanmoy Dam, Mahardhika Pratama, MD Meftahul Ferdaus, Sreenatha Anavatti and Hussein Abbas
Requirements
- Pytorch 1.10.0
- CUDA 11.4
All the experiments were tested on a single NVIDIA RTX 3080 GPU with 16Gb memory.
Benchmarks
- Prepare Data