Michael Beukman
YOU?
Author Swipe
View article: An Optimisation Framework for Unsupervised Environment Design
An Optimisation Framework for Unsupervised Environment Design Open
For reinforcement learning agents to be deployed in high-risk settings, they must achieve a high level of robustness to unfamiliar scenarios. One method for improving robustness is unsupervised environment design (UED), a suite of methods …
View article: Kinetix: Investigating the Training of General Agents through Open-Ended Physics-Based Control Tasks
Kinetix: Investigating the Training of General Agents through Open-Ended Physics-Based Control Tasks Open
While large models trained with self-supervised learning on offline datasets have shown remarkable capabilities in text and image domains, achieving the same generalisation for agents that act in sequential decision problems remains an ope…
View article: JaxLife: An Open-Ended Agentic Simulator
JaxLife: An Open-Ended Agentic Simulator Open
Human intelligence emerged through the process of natural selection and evolution on Earth. We investigate what it would take to re-create this process in silico. While past work has often focused on low-level processes (such as simulating…
View article: RobocupGym: A challenging continuous control benchmark in Robocup
RobocupGym: A challenging continuous control benchmark in Robocup Open
Reinforcement learning (RL) has progressed substantially over the past decade, with much of this progress being driven by benchmarks. Many benchmarks are focused on video or board games, and a large number of robotics benchmarks lack diver…
View article: JaxUED: A simple and useable UED library in Jax
JaxUED: A simple and useable UED library in Jax Open
We present JaxUED, an open-source library providing minimal dependency implementations of modern Unsupervised Environment Design (UED) algorithms in Jax. JaxUED leverages hardware acceleration to obtain on the order of 100x speedups compar…
View article: Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement Learning
Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement Learning Open
Benchmarks play a crucial role in the development and analysis of reinforcement learning (RL) algorithms. We identify that existing benchmarks used for research into open-ended learning fall into one of two categories. Either they are too …
View article: Refining Minimax Regret for Unsupervised Environment Design
Refining Minimax Regret for Unsupervised Environment Design Open
In unsupervised environment design, reinforcement learning agents are trained on environment configurations (levels) generated by an adversary that maximises some objective. Regret is a commonly used objective that theoretically results in…
View article: Dynamics Generalisation in Reinforcement Learning via Adaptive Context-Aware Policies
Dynamics Generalisation in Reinforcement Learning via Adaptive Context-Aware Policies Open
While reinforcement learning has achieved remarkable successes in several domains, its real-world application is limited due to many methods failing to generalise to unfamiliar conditions. In this work, we consider the problem of generalis…
View article: Hierarchical WaveFunction Collapse
Hierarchical WaveFunction Collapse Open
Video game developers are increasingly utilising procedural content generation (PCG) techniques in order to generate more content far quicker than if it were designed. Although promising, much of the successful work to date has been achiev…
View article: Analysing Cross-Lingual Transfer in Low-Resourced African Named Entity Recognition
Analysing Cross-Lingual Transfer in Low-Resourced African Named Entity Recognition Open
Transfer learning has led to large gains in performance for nearly all NLP tasks while making downstream models easier and faster to train. This has also been extended to low-resourced languages, with some success. We investigate the prope…
View article: Hierarchically Composing Level Generators for the Creation of Complex Structures
Hierarchically Composing Level Generators for the Creation of Complex Structures Open
Procedural content generation (PCG) is a growing field, with numerous applications in the video game industry and great potential to help create better games at a fraction of the cost of manual creation. However, much of the work in PCG is…
View article: Analysing Cross-Lingual Transfer in Low-Resourced African Named Entity Recognition
Analysing Cross-Lingual Transfer in Low-Resourced African Named Entity Recognition Open
Michael Beukman, Manuel Fokam. Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers…
View article: MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition
MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition Open
African languages are spoken by over a billion people, but are underrepresented in NLP research and development. The challenges impeding progress include the limited availability of annotated datasets, as well as a lack of understanding of…
View article: Augmentative Topology Agents For Open-Ended Learning
Augmentative Topology Agents For Open-Ended Learning Open
In this work, we tackle the problem of open-ended learning by introducing a method that simultaneously evolves agents and increasingly challenging environments. Unlike previous open-ended approaches that optimize agents using a fixed neura…
View article: Separating Grains from the Chaff: Using Data Filtering to Improve Multilingual Translation for Low-Resourced African Languages
Separating Grains from the Chaff: Using Data Filtering to Improve Multilingual Translation for Low-Resourced African Languages Open
We participated in the WMT 2022 Large-Scale Machine Translation Evaluation for the African Languages Shared Task. This work describes our approach, which is based on filtering the given noisy data using a sentence-pair classifier that was …
View article: Procedural content generation using neuroevolution and novelty search for diverse video game levels
Procedural content generation using neuroevolution and novelty search for diverse video game levels Open
Procedurally generated video game content has the potential to drastically\nreduce the content creation budget of game developers and large studios.\nHowever, adoption is hindered by limitations such as slow generation, as well\nas low qua…
View article: Adaptive Online Value Function Approximation with Wavelets
Adaptive Online Value Function Approximation with Wavelets Open
Using function approximation to represent a value function is necessary for continuous and high-dimensional state spaces. Linear function approximation has desirable theoretical guarantees and often requires less compute and samples than n…
View article: Towards Objective Metrics for Procedurally Generated Video Game Levels
Towards Objective Metrics for Procedurally Generated Video Game Levels Open
With increasing interest in procedural content generation by academia and game developers alike, it is vital that different approaches can be compared fairly. However, evaluating procedurally generated video game levels is often difficult,…
View article: A Few Thousand Translations Go a Long Way! Leveraging Pre-trained Models for African News Translation
A Few Thousand Translations Go a Long Way! Leveraging Pre-trained Models for African News Translation Open
David Adelani, Jesujoba Alabi, Angela Fan, Julia Kreutzer, Xiaoyu Shen, Machel Reid, Dana Ruiter, Dietrich Klakow, Peter Nabende, Ernie Chang, Tajuddeen Gwadabe, Freshia Sackey, Bonaventure F. P. Dossou, Chris Emezue, Colin Leong, Michael …
View article: MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition
MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition Open
David Adelani, Graham Neubig, Sebastian Ruder, Shruti Rijhwani, Michael Beukman, Chester Palen-Michel, Constantine Lignos, Jesujoba Alabi, Shamsuddeen Muhammad, Peter Nabende, Cheikh M. Bamba Dione, Andiswa Bukula, Rooweither Mabuya, Bonav…