Kale-ab Tessera

PhD Candidate

Autonomous Agents Group, University of Edinburgh

Biography

Kale-ab is a PhD candidate in the Autonomous Agents research group at the University of Edinburgh, supervised by Dr. Stefano Albrecht. His research focuses on developing Multi-Agent Reinforcement Learning (MARL) systems for effective real-world collaboration. Before his PhD, he gained 4.5 years of experience in machine learning, including 2.5 years as a research engineer at InstaDeep, and 3 years of experience in software engineering. Kale-ab is also committed to supporting impactful tech projects in Africa and promoting diversity within the machine learning community.

For more information, you can view my resumé .

Interests

Multi-Agent Reinforcement Learning (MARL)
Reinforcement Learning
Deep Learning

Education

MSc in Computer Science, Focusing on Deep Learning (Distinction), 2018 - March, 2021
University of Witwatersrand
Honours in Computer Science (Distinction), 2016
University of Pretoria
BSc. in Computer Science, 2013 - 2015
University of Pretoria

Recent Publications

Omayma Mahjoub, Ruan John de Kock, Siddarth Singh, Wiem Khlifi, Abidine Vall, Kale-ab Tessera, Rihab Gorsane, Arnu Pretorius (2024). Efficiently Quantifying Individual Agent Importance in Cooperative MARL. (Oral) eXplainable AI approaches for deep reinforcement learning (XAI4DRL) Workshop @ AAAI, 2024.

PDF

Siddarth Singh, Omayma Mahjoub, Ruan John de Kock, Wiem Khlifi, Abidine Vall, Kale-ab Tessera, Arnu Pretorius (2024). How much can change in a year? Revisiting Evaluation in Multi-Agent Reinforcement Learning. eXplainable AI approaches for deep reinforcement learning (XAI4DRL) Workshop @ AAAI, 2024.

PDF

Kale-ab Tessera, Callum Tilbury, Sasha Abramowitz, Ruan de Kock, Omayma Mahjoub, Benjamin Rosman, Sara Hooker, Arnu Pretorius (2023). Generalisable Agents for Neural Network Optimisation. WANT@ NeurIPS 2023 and OPT NeurIPS 2023 workshops.

PDF Poster

Andries Smit, Paul Duckworth, Nathan Grinsztajn, Kale-ab Tessera, Thomas Barrett, Arnu Pretorius (2023). Are we going MAD? Benchmarking Multi-Agent Debate between Language Models for Medical Q&A. Deep Generative Models for Health Workshop NeurIPS 2023.

PDF

Claude Formanek, Callum Rhys Tilbury, Jonathan Shock, Kale-ab Tessera, Arnu Pretorius (2023). Reduce, Reuse, Recycle: Selective Reincarnation in Multi-Agent Reinforcement Learning. (Oral) Workshop on Reincarnating Reinforcement Learning at ICLR 2023.

PDF

Kale-ab Tessera, Chiratidzo Matowe, Arnu Pretorius, Benjamin Rosman, Sara Hooker (2022). Just-in-Time Sparsity: Learning Dynamic Sparsity Schedules. Dynamic Neural Networks ICML Workshop 2022.

PDF Poster Slides

Ibrahim Salihu Yusuf, Kale-ab Tessera, Thomas Tumiel, Zohra Slim, Amine Kerkeni, Sella Nevo, Arnu Pretorius (2021). On pseudo-absence generation and machine learning for locust breeding ground prediction in Africa. AI + HADR 2021 and ML4D 2021 NeurIPS Workshops.

PDF Cite Code Blog

Arnu Pretorius, Kale-ab Tessera, Andries P. Smit, Claude Formanek, St John Grimbly, Kevin Eloff, Siphelele Danisa, Lawrence Francis, Jonathan Shock, Herman Kamper, Willie Brink, Herman Engelbrecht, Alexandre Laterre, Karim Beguir (2021). Mava: a research framework for distributed multi-agent reinforcement learning.

PDF Cite Code Blog

Kale-ab Tessera, Sara Hooker, Benjamin Rosman (2021). Keep the Gradients Flowing: Using Gradient Flow to Study Sparse Network Optimization. Sparsity in Neural Networks Workshop 2021.

PDF Cite Poster Slides

Kale-ab Tessera (2021). On Sparsity In Deep Learning: The Benefits And Pitfalls Of Sparse Neural Networks And How To Learn Their Architectures. Faculty of Science, University of the Witwatersrand.

PDF

Recent & Upcoming Talks

Introduction to Deep Reinforcement Learning - University of Pretoria

May 26, 2023 5:00 PM

Introduction to Deep Reinforcement Learning - Indaba X Ghana

Sep 23, 2022 5:00 PM

“Introduction to ML using JAX” and “Introduction to Reinforcement Learning” tutorials

Aug 25, 2022 8:00 AM

Spotlight talk - Indaba 2022

Aug 25, 2022 8:00 AM

Playing Starcraft and Saving the World Using Multi-agent Reinforcement Learning

Oct 29, 2021 5:00 PM

See all events

Open Source

You can see a full list of my open source projects on github - https://github.com/KaleabTessera .

Mava: a research framework for distributed multi-agent reinforcement learning

A framework for MARL, written in JAX.

Deep Learning Indaba Practicals 2022

A collection of high-quality practicals covering a variety of modern machine learning techniques.

Pseudo Absence Generation and Locust Prediction

Locust breeding ground prediction using pseudo-absence generation and machine learning.

Baobab

Baobab is an open source multi-tenant web application designed to facilitate the application and selection process for large scale meetings within the machine learning and artificial intelligence communities globally.