Excited to be at @RL_Conference 🇨🇦 this week! I'm presenting two talks on logical composition in RL (Aug 6, Track 3) + a poster at @RLFrameWorkshop on a simple yet effective approach for non-Markov RL (Aug 5). Huge thanks to my amazing co-authors 🥳. Come say hi!
Thrilled to share that our paper was just published in JAIR (https://t.co/awRLUYc7YH)! We formalise task composition in RL using lattice structures📷, building a general framework for logical composition over arbitrary tasks beyond Boolean logic. (1/8)
🧵👇
🌟Celebrating African research excellence at #DLI2025! 🌟
The Kambule Doctoral Award recognises and encourages excellence in research and writing by doctoral candidates at African universities, in any area of computational and statistical sciences. This award celebrates African research excellence and honours Thamsanqa Kambule’s legacy as a defender of learning, seeker of knowledge, and activist for equality.
👏 Congratulations to Lexy Andati (Rhodes University & SARAO) and Geraud Nangue Tasse ( University of the Witwatersrand).
#DLI2025 #Indaba2025 #Urunana
🚀 We’re excited to unveil the first lineup of technical tutorials and practicals at #DLI2025!
These sessions go beyond training—they reflect our mission of building AI in Africa, “Urunana” (hand in hand). Attendees will gain hands-on experience, exploring advanced topics:
Pracicals:
✨ Machine Learning Foundations
✨ Generative Models & LLMs for African languages
Tutorials:
✨ Mathematics of Deep Learning
✨ Introduction to Deep Generative Models
📌 All tutorial content will also be available online after the Indaba. Don’t miss out—subscribe here 👉 https://t.co/RzFnqGMVRP
Excited to be at @RL_Conference 🇨🇦 this week! I'm presenting two talks on logical composition in RL (Aug 6, Track 3) + a poster at @RLFrameWorkshop on a simple yet effective approach for non-Markov RL (Aug 5). Huge thanks to my amazing co-authors 🥳. Come say hi!
Thrilled to share that our paper was just published in JAIR (https://t.co/awRLUYc7YH)! We formalise task composition in RL using lattice structures📷, building a general framework for logical composition over arbitrary tasks beyond Boolean logic. (1/8)
🧵👇
Our paper, “Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks” will be presented at ICLR 2025 this week (https://t.co/coGozhcYmZ)! We derive closed-form dynamics for some, remarkably linear, feature learning ReLU networks (1/9)
Many thanks to my supervisors Steven James & @BenjaminRosman to their extensive help in the long journey to this paper. If you're working on: multitask RL, interpretable RL, logical reasoning in RL, sample-efficiency and generalisation ... this might be of interest! (8/8)
Thrilled to share that our paper was just published in JAIR (https://t.co/awRLUYc7YH)! We formalise task composition in RL using lattice structures📷, building a general framework for logical composition over arbitrary tasks beyond Boolean logic. (1/8)
🧵👇
New Article: "Composition and Zero-Shot Transfer with Lattice Structures in Reinforcement Learning" by Nangue Tasse, James, and Rosman https://t.co/QbFIpusyKa
This work builds on our Boolean Task Algebra NeurIPS paper, but takes it much further: E.g. Generalises:
Task composition to any environment (even stochastic) and any reward function; Extended value functions to WVFs, supporting general deterministic goal-reaching tasks. (7/8)