Zhengyao Jiang

icon indicating a website link zhengyaojiang.github.io icon indicating an email [email protected]

icon company University College London icon location London, UK icon location PhD student at UCL, Interested in Offline Reinforcement Learning (RL), Data-Efficient RL and Neuro-Symbolic Methods for RL.

Results 4 comments of Zhengyao Jiang

adding new layers and supporting multiple pre-configured networks in net config

Thanks for your contribution again! > Adding support for additional layers such as Batch Normalization and ReLU Yes, this makes sense. > Support for multiple pre-configured networks in net_config.json. For...

Low benefit if the end_date is after 2018/02/10

We have two guesses about this: 1. the agent perform worse when there are big differences in the market. The agent bases its decision on the history of the market....

About the volume feature

> It seems the framework does not support volume feature. Is there any reason that the framework can not support volume feature? I think volume feature is very important to...

Loss function only contains instantaneous reward but not cumulated reward

> However, it seems that such loss function only contains instantaneous reward, not average cumulated reward. If there is no commission fee, when the action won't affect the state transition,...