ContinualAI Seminar: "S-IL: Natural continual learning: success is a journey, not (just) a destination"
Abstract: Biological agents are known to learn many different tasks over the course of their lives, and to be able to revisit previous tasks and behaviors with little to no loss in performance. In contrast, artificial agents are prone to 'catastrophic forgetting', whereby performance on previous tasks deteriorates rapidly as new ones are acquired. This shortcoming has recently been addressed using methods that encourage parameters to stay close to those used for previous tasks. This can be done by (i) using specific parameter regularizers that map out suitable destinations in parameter space, or (ii) guiding the optimization journey by projecting gradients into subspaces that do not interfere with previous tasks. However, these methods often exhibit subpar performance in both feedforward and recurrent neural networks, with recurrent networks being of interest to the study of neural dynamics supporting biological continual learning. In this work, we propose Natural Continual Learning (NCL), a new method that unifies weight regularization and projected gradient descent. NCL uses Bayesian weight regularization to encourage good performance on all tasks at convergence and combines this with gradient projection using the prior precision, which prevents catastrophic forgetting during optimization. Our method outperforms both standard weight regularization techniques and projection based approaches when applied to continual learning problems in feedforward and recurrent networks. Finally, the trained networks evolve task-specific dynamics that are strongly preserved as new tasks are learned, similar to experimental findings in biological circuits.
You can discuss about this seminar in our forum: [ Ссылка ]
------------------------ About ContinualAI ------------------------
ContinualAI is an official non-profit research organization and the largest open community on Continual Learning for AI. We aim at connecting people and working better together on this fascinating topic we consider fundamental for the future of AI.
• Official website: [ Ссылка ]
• Join us now at [ Ссылка ]
Please consider supporting us with a small donation at: [ Ссылка ]
It's thanks to people like you that we are making ContinualAI a reality!
