Great work by Aswin Paul reducing the computational complexity of sophisticated #active #inference by orders of magnitude:

📢Our latest preprint reveals that biological neurons can learn faster than deep Reinforcement Learning algorithms. Using DishBrain, we compared in vitro neuronal networks with RL algorithms. The result? Biological neurons showed superior sample efficiency and learning speed!

The amazing team who made this possible: Moein Khajehnejad Aswin Paul Adeel Razi Brett J. Kagan Cortical Labs arxiv.org/abs/2405.16946 #AI #NeuroAI #ReinforcementLearning #MachineLearning #Neuroscience #SampleEfficiency

Amazing work from Cortical Labs and Monash Biomedical Imaging. Many more to come!

1/8 📢Our latest preprint, "Biological Neurons Compete with Deep Reinforcement Learning in Sample Efficiency in a Simulated Gameworld," is now available! 📄 Check it out [arxiv.org/pdf/2405.16946]. Read on for a brief overview!

8/8 Special thanks to my co-authors Forough Habibollahi , Aswin Paul , Adeel Razi , and Brett J. Kagan . Your collaboration made this possible! 👏 Turner Institute for Brain and Mental Health Cortical Labs

Aswin is putting paper after paper from his PhD thesis. Have a look at the thread. Awesome collaboration with Takuya Isomura at RIKEN Center for Brain Science

It was such a great symposium on naturalistic imaging with Emily Finn, The LAB Lab 🏳️‍🌈🏳️‍⚧️, Hongmi Lee, Asirh Zadbood, together with Sadia Shakil at #OHBM2024

Thread with a few posts and papers about active inference or free energy principle. 1/ x.com/aswinpaule/sta…

Very grateful and excited to be promoted to Professor of Computational Neuroscience at Monash. A big shout out to all my mentors, trainees, colleagues and collaborators, without their help it most definitely wasn’t possible. Thank you so very much 🙏

Excited to share the first paper of my PhD: Towards a theory of learning dynamics in deep state space models arxiv.org/abs/2407.07279 with Jimmy Smith, Michael Kleinman, Dan Biderman, and Scott Linderman. Accepted as a Spotlight talk at the NGSM workshop at ICML 2024!

📢We're delighted to organise NeuroAI workshop at this year's NeurIPS Conference 2024 (excuse the typo in the graphic). Accepting submissions (4 pages + supp): neuroai-workshop.github.io Highlight: In-person panel with Yoshua Bengio and Karl Friston Follow NeuroAI Workshop @ NeurIPS2024 for details

📢 Attention! Pleased to announce we are organizing the 1st NeuroAI workshop NeurIPS Conference 2024! Deadline for submissions: 7th September More details: neuroai-workshop.github.io Follow NeuroAI Workshop @ NeurIPS2024 Forough Habibollahi Adeel Razi Noor Sajid Jason Eshraghian Tony Zador Yoshua Bengio

Article | From pixels to planning: scale-free active inference Coauthors: Conor Heins, Tim Verbelen, Lancelot Da Costa, Tommaso Salvatori, Dimitrije Markovic, Alexander Tschantz, Magnus Koudahl, Chris L Buckley, Thomas Parr arxiv.org/pdf/2407.20292

📢Call for abstracts Joint Meeting of OHBM Australia + `Maths in the Brain', 7-8 Nov, 2024, in Melbourne. Abstract submission: tinyurl.com/486c9t8w Deadline: 13 September 2024. Registration here: tinyurl.com/jvwcmcte For more information: tinyurl.com/yubdarc2

Model based inference and control is the future of AI ❤️

We didn’t

Exciting news! Nature Machine Intelligence has featured our work on decentralized multi-agent reinforcement learning for large-scale networked control, on cracking the scaling challenge of #MARL. The article is open access: nature.com/articles/s4225… Nature Portfolio Nature Machine Intelligence