Learning to Represent Action Values as a Hypergraph on the Action Vertices

Authors: Arash Tavakoli, Mehdi Fatemi, Petar Kormushev

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We show the effectiveness of our approach on a myriad of domains: illustrative prediction problems under minimal confounding effects, Atari 2600 games, and discretised physical control benchmarks.
Researcher Affiliation Collaboration Arash Tavakoli Imperial College London Mehdi Fatemi Microsoft Research Montréal Petar Kormushev Imperial College London
Pseudocode No The paper does not contain any pseudocode or algorithm blocks.
Open Source Code Yes Code availability The source code can be accessed at: https://github.com/atavakol/ action-hypergraph-networks.
Open Datasets Yes We tested our approach in Atari 2600 games using the Arcade Learning Environment (ALE) (Bellemare et al., 2013). ... To test our approach in problems with larger action spaces, we consider a suite of physical control benchmarks simulated using Py Bullet (Coumans & Bai, 2019).
Dataset Splits No The paper refers to training and testing processes but does not explicitly provide details on validation splits or specific train/test/validation data partitioning proportions or methods within the main text or appendices.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU, GPU models, or memory specifications) used for running the experiments.
Software Dependencies No The paper mentions 'TensorFlow (Abadi et al., 2016) and NumPy (Harris et al., 2020)' but does not specify version numbers for these software components.
Experiment Setup Yes Table 2: Hyperparameters used for HGQN and DQN in our physical control benchmarks. This table lists specific values for minibatch size, replay memory size, learning rate, optimiser (Adam), etc.