Expert-augmented actor-critic for ViZDoom and Montezumas Revenge
By Michal Garmulewicz, Henryk Michalewski, Piotr Milo's. 2018
8 citations recorded by the source corpus.
We propose an expert-augmented actor-critic algorithm, which we evaluate on two environments with sparse rewards: Montezumas Revenge and a demanding maze from the ViZDoom suite. In the case of Montezumas Revenge, an agent trained with our method achieves very good results consistently scoring above 27,000 points (in many experiments beating the first world). With an appropriate choice of hyperparameters, our algorithm surpasses the performance of the expert data. In a number of experiments, we have observed an unreported bug in Montezumas Revenge which allowed the agent to score more than 800,000 points.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sergey Levine
- Amos J. Storkey
- Harrison Edwards
- Chelsea Finn
- Yuri Burda
- Oleg Klimov
- Ruslan Salakhutdinov
- Pieter Abbeel
Citations in the corpus
- Exploration by Random Network Distillation (2018, methodology): “Significant progress has been achieved by methods with access to either expert demonstrations (Pohlen et al., 2018; Aytar et al., 2018; Garmulewicz et al., 2018), special access to the underlying emulator state (Tang et al.”