Slow and Steady Wins the Race: Maintaining Plasticity with Hare and Tortoise Networks
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909475269509120 |
|---|---|
| author | Lee, Hojoon Cho, Hyeonseo Kim, Hyunseung Kim, Donghu Min, Dugki Choo, Jaegul Lyle, Clare |
| author_facet | Lee, Hojoon Cho, Hyeonseo Kim, Hyunseung Kim, Donghu Min, Dugki Choo, Jaegul Lyle, Clare |
| contents | This study investigates the loss of generalization ability in neural networks, revisiting warm-starting experiments from Ash & Adams. Our empirical analysis reveals that common methods designed to enhance plasticity by maintaining trainability provide limited benefits to generalization. While reinitializing the network can be effective, it also risks losing valuable prior knowledge. To this end, we introduce the Hare & Tortoise, inspired by the brain's complementary learning system. Hare & Tortoise consists of two components: the Hare network, which rapidly adapts to new information analogously to the hippocampus, and the Tortoise network, which gradually integrates knowledge akin to the neocortex. By periodically reinitializing the Hare network to the Tortoise's weights, our method preserves plasticity while retaining general knowledge. Hare & Tortoise can effectively maintain the network's ability to generalize, which improves advanced reinforcement learning algorithms on the Atari-100k benchmark. The code is available at https://github.com/dojeon-ai/hare-tortoise. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_02596 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Slow and Steady Wins the Race: Maintaining Plasticity with Hare and Tortoise Networks Lee, Hojoon Cho, Hyeonseo Kim, Hyunseung Kim, Donghu Min, Dugki Choo, Jaegul Lyle, Clare Machine Learning Artificial Intelligence This study investigates the loss of generalization ability in neural networks, revisiting warm-starting experiments from Ash & Adams. Our empirical analysis reveals that common methods designed to enhance plasticity by maintaining trainability provide limited benefits to generalization. While reinitializing the network can be effective, it also risks losing valuable prior knowledge. To this end, we introduce the Hare & Tortoise, inspired by the brain's complementary learning system. Hare & Tortoise consists of two components: the Hare network, which rapidly adapts to new information analogously to the hippocampus, and the Tortoise network, which gradually integrates knowledge akin to the neocortex. By periodically reinitializing the Hare network to the Tortoise's weights, our method preserves plasticity while retaining general knowledge. Hare & Tortoise can effectively maintain the network's ability to generalize, which improves advanced reinforcement learning algorithms on the Atari-100k benchmark. The code is available at https://github.com/dojeon-ai/hare-tortoise. |
| title | Slow and Steady Wins the Race: Maintaining Plasticity with Hare and Tortoise Networks |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2406.02596 |