Residual CNN
A convolutional neural network that mitigates the vanishing gradient problem in deep networks through skip connections. It is used as the actor-critic network to extract spatial features of the board.
Timeline (last 90 days)
The training configuration used the PPO + GAE algorithm, 32 parallel environments, a rollout length of 256, gamma 0.998, lambda 0.97, and a small residual CNN as the actor-critic.
The trajectory curvature effect persists under matched performance and controlled training duration, and generalizes from synthetic vector inputs to the training of residual CNNs on rendered images and natural images.
All Facts (2)
The training configuration used the PPO + GAE algorithm, 32 parallel environments, a rollout length of 256, gamma 0.998, lambda 0.97, and a small residual CNN as the actor-critic.
50%UnverifiedThe trajectory curvature effect persists under matched performance and controlled training duration, and generalizes from synthetic vector inputs to the training of residual CNNs on rendered images and natural images.
50%