maml

Tag

Cards List
#maml

On first-order meta-learning algorithms

OpenAI Blog · 2018-03-08 Cached

This paper analyzes first-order meta-learning algorithms for few-shot learning, introducing Reptile and providing theoretical insights into why these computationally efficient methods work well on established benchmarks.

0 favorites 0 likes
#maml

Reptile: A scalable meta-learning algorithm

OpenAI Blog · 2018-03-07 Cached

OpenAI introduces Reptile, a scalable meta-learning algorithm for few-shot classification that achieves comparable performance to MAML while converging faster with lower variance. The paper provides theoretical analysis showing Reptile maximizes inner product between task gradients for improved generalization.

0 favorites 0 likes
#maml

Some considerations on learning to explore via meta-reinforcement learning

OpenAI Blog · 2018-03-03 Cached

OpenAI researchers introduce E-MAML and E-RL², two meta-reinforcement learning algorithms designed to improve exploration in tasks where discovering optimal policies requires significant exploration. The work demonstrates these algorithms' effectiveness on novel environments including Krazy World and maze tasks.

0 favorites 0 likes
← Back to home

Submit Feedback