Ryan Kortvelesy
Ryan Kortvelesy
Home
Publications
Posts
Talks
Contact
Light
Dark
Automatic
paper-conference
Generalised $f$-Mean Aggregation for Graph Neural Networks
A generalised learnable aggregator that can boost the representational complexity of Graph Neural Networks.
Ryan Kortvelesy
,
Steven Morad
,
Amanda Prorok
PDF
Cite
Code
Slides
Reinforcement Learning with Fast and Forgetful Memory
A fast memory model for reinforcement learning.
Steven Morad
,
Ryan Kortvelesy
,
Stephan Liwicki
,
Amanda Prorok
PDF
Cite
Code
Slides
POPGym: Benchmarking Partially Observable Reinforcement Learning
A suite of partially observable gym environments for benchmarking memory models in reinforcement learning.
Steven Morad
,
Ryan Kortvelesy
,
Matteo Bettini
,
Stephan Liwicki
,
Amanda Prorok
PDF
Cite
Code
Permutation-Invariant Set Autoencoders with Fixed-Size Embeddings for Multi-Agent Learning
A set autoencoder that defines a bijective mapping between variable-sized sets and fixed-sized embeddings.
Ryan Kortvelesy
,
Steven Morad
,
Amanda Prorok
PDF
Cite
Code
Fixed Integral Neural Networks
A method for analytically computing integrals over neural networks.
Ryan Kortvelesy
PDF
Cite
Code
VMAS: A Vectorized Multi-Agent Simulator for Collective Robot Learning
A vectorised simulation environment for reinforcement learning.
Matteo Bettini
,
Ryan Kortvelesy
,
Jan Blumenkamp
,
Amanda Prorok
PDF
Cite
Code
Video
QGNN: Value Function Factorisation with Graph Neural Networks
A GNN-based value factorisation method, enabling credit assignment on collaborative tasks (
i.e.
tasks with interdependencies between agents).
Ryan Kortvelesy
,
Amanda Prorok
PDF
Cite
An example conference paper
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis posuere tellus ac convallis placerat. Proin tincidunt magna sed ex sollicitudin condimentum.
Ryan Kortvelesy
,
Robert Ford
Jul 1, 2013
PDF
Cite
Code
Dataset
Video
Source Document
Cite
×