Alex Nichol
Alex Nichol
Previously, the BigGAN-deep discriminator applied a ReLU to the input images, which range in [-1, 1]. This appeared to cripple the training process.
I am in the process of figuring out how my microkernel should implement inter-process communication. Here are the two types of IPC I want to consider: - Shared memory IPC...
This way, you can use a Roller in conjunction with DQN (instead of needing to use a Player).
This may make it possible to have more optimized backprop.
In particular, it would be nice to turn the existing docstrings, classes, and functions into a browsable web-friendly format.
Right now, there's three different types a model state can be: `NoneType`, an array, or a tuple of arrays. In the future, it would be nice to unify this. I...
This is a "nice to have", but not a necessity.
[Distributional Reinforcement Learning with Quantile Regression](https://arxiv.org/abs/1710.10044) This supposedly improves the performance of distributional DQN, so I'd like to add it as an option for distributional Q-networks.