1 citations · 2 across the 9 of their papers we have counts for
11 papers
Local Updates, Global Learning (LUGL): Playing Games with non-incremental Learners
David Milec, Spyridon Samothrakis, Michael Fairbank +1
The dominance of Neural Networks (NNs) in RL is partially due to their incremental learning capability, which naturally suits the online, non-stationary nature of self-play trainin…
Disassociating performance from compositional feature learning
George Dimitriadis, Spyridon Samothrakis
Out-of-distribution (OOD) generalisation through composition requires a system to discover invariant properties from input-output associations and transfer them to novel inputs and…
A Research Agenda for Usability and Generalisation in Reinforcement Learning
Dennis J. N. J. Soemers, Spyridon Samothrakis, Kurt Driessens +1
It is common practice in reinforcement learning (RL) research to train and deploy agents in bespoke simulators, typically implemented by engineers directly in general-purpose progr…
Games of Knightian Uncertainty as AGI testbeds
Spyridon Samothrakis, Dennis J. N. J. Soemers, Damian Machlanski
Arguably, for the latter part of the late 20th and early 21st centuries, games have been seen as the drosophila of AI. Games are a set of exciting testbeds, whose solutions (in ter…
Robustness of Algorithms for Causal Structure Learning to Hyperparameter Choice
Damian Machlanski, Spyridon Samothrakis, Paul Clarke
Hyperparameters play a critical role in machine learning. Hyperparameter tuning can make the difference between state-of-the-art and poor prediction performance for any algorithm,…
The Challenges of Hyperparameter Tuning for Accurate Causal Effect Estimation
Damian Machlanski, Spyridon Samothrakis, Paul Clarke
ML is playing an increasingly crucial role in estimating causal effects of treatments on outcomes from observational data. Many ML methods (`causal estimators') have been proposed…
Undersmoothing Causal Estimators with Generative Trees
Damian Machlanski, Spyros Samothrakis, Paul Clarke
Inferring individualised treatment effects from observational data can unlock the potential for targeted interventions. It is, however, hard to infer these effects from observation…
A Comparison of Self-Play Algorithms Under a Generalized Framework
Daniel Hernandez, Kevin Denamganai, Sam Devlin +2
Throughout scientific history, overarching theoretical frameworks have allowed researchers to grow beyond personal intuitions and culturally biased theories. They allow to verify a…
Deep Learning in Target Space
Michael Fairbank, Spyridon Samothrakis, Luca Citi
Deep learning uses neural networks which are parameterised by their weights. The neural networks are usually trained by tuning the weights to directly minimise a given loss functio…
Open Loop In Natura Economic Planning
Spyridon Samothrakis
The debate between the optimal way of allocating societal surplus (i.e. products and services) has been raging, in one form or another, practically forever; following the collapse…
Viewpoint: Artificial Intelligence and Labour
Spyridon Samothrakis
The welfare of modern societies has been intrinsically linked to wage labour. With some exceptions, the modern human has to sell her labour-power to be able reproduce biologically…