Termsslug: weight-perturbation

Weight Perturbation

权重扰动

A gradient-free or approximate-gradient learning method that perturbs network weights and selects updates by reward.

Curated Intelligence

1 items · Page 1/1
Hacker News AIT2·Models
42 pts
A Visual Explainer of Reinforcement Learning and Weight Perturbation

This is an interactive popular-science explainer that visualizes the fundamentals of reinforcement learning, using Braitenberg vehicles to show how neural networks can learn via weight perturbation instead of backpropagation. It notes that OpenAI used a similar evolution-strategies approach in 2017 to train AI to play Atari games. The piece is educational and contains no new research results or model releases.