Open Call for Papers for 2022 ICLR Workshop on Collective Learning
The 2022 workshop, “From Cells to Societies: Collective Learning Across Scales,” will examine natural systems to understand how collective interactions can contribute to new learning approaches in artificial learning systems.
An Open-ended Environment for Lifelong Learning AI
Associate Professor Julian Togelius is working on a scalable generative AI benchmark that will allow for the development of open-ended learning, task transfer, and generalization.
GPU Grant to Support TU Delft’s AI Research
Wendelin Böhmer investigates how restricted neural modules, similar to Badger agents, can learn generalizing sub-tasks and how a message passing architecture can reliably compose their output for out-of-distribution generalization.
Bridging the Continual Learning Gap with Deep Artificial Neurons
Integrating GoodAI’s Badger architecture and principles, new grant recipients Rolando Estrada and Blake Camp aim to develop a continual semi-supervised learning system capable of joint supervised and unsupervised learning.
Memetic Experiments for an Open-Ended Learning Paradigm
GoodAI presents a new paper to trace memetic evolution and investigate whether open-ended learning algorithms can be discovered through the lens of socio-cultural learning models.
Badger Seminar Summer 2021 Summary: Beyond Life-long Learning via Modular Meta-Learning
We recently held the Badger Seminar titled: Beyond Life-long Learning via Modular Meta-Learning, with participants joining online and at our headquarters in Prague.
ALIFE 2021: celebrating the science and art of artificial life
Last week saw the second fully virtual ALIFE conference bringing together nearly 400 leading researchers and practitioners from across the world working on problems related to simulating artificial life.
GoodAI Scientist Nicholas Guttenberg wins the Evocraft Minecraft Open-Endedness Challenge
We are very proud that Nicholas Guttenberg, Senior Research Scientist at GoodAI, recently won the Evocraft 2021 competition - the Minecraft Open-Endedness Challenge!
GoodAI Research Roadmap 2021/2022
In 2021/22 GoodAI will focus on four core research areas: learning to learn, lifelong (gradual) learning, open-endedness, and generalization / extrapolation of meta-learned algorithms.
Building a research culture where we share questions and difficulties, not just answers
Is sharing where we’re stuck more important to furthering research than sharing the answers we’ve found?