ogmacorp / ogmaneo2 Goto Github PK

I just started looking at the OgmaNeo2 implementation and trying to understand how it works. Within the draft paper, you mention 2 algorithms to encode 2SI and ESR. While going through the code, I only see an implementation of a ImageEncoder. Is that using ESR, 2SI or a different algorithm

Project looks great!

The presentation link goes to a 404

Reverting to the old pointer-shift buffer improves performance for some reason, although it should be equivalent.

I have tested the new encoder on CartPole-v1 and MiniWorld-Hallway-v0 (from gym_miniworld). The performance is significantly worse, where in CartPole-v1 for example the agent achieves only a slight improvement in 1000 episodes, in comparison to the old encoder, where the agent is able to achieve 500 timestep episodes almost every time in CartPole-v1 after 1000 episodes. The same worsening of performance can be seen in MiniWorld-Hallway-v0. Can you replicate these findings? (For MiniWorld-Hallway-v0 I used layerSizes=6*[Int3(6,6,32)])

I unfortunately can not state the exact reason for this worsening, since I don't fully understand the new encoder. About that, I also have a question not related to this issue: Would it be beneficial to introduce a decorrelation step into the encoders? This should ensure that different columns don't encode the same information which should lead to a higher quality encoding.