Artificial Intelligence: What It Really Is (and Isn’t)
We had the pleasure of speaking with Prof. Michael Littman (Brown University) about the fundamentals of AI and reinforcement learning. We explore whether today’s AI systems are truly “intelligent”—or just faking it impressively well.
Listen here: https://youtu.be/N3TpwsMVeRg
How a big shift in training LLMs led to a capability explosion https://arstechni.ca/FerG #reinforcementlearning #imitationlearning #explainers #Features #AI
Good article how reinforcement learning improved current AI models. Also illustrates that LLMs today are not just imitating.
https://arstechnica.com/ai/2025/07/how-a-big-shift-in-training-llms-led-to-a-capability-explosion/?utm_brand=arstechnica&utm_social-type=owned&utm_source=mastodon&utm_medium=social
#AI #reinforcementlearning
"Intelligence is figuring out how the world works rather than waiting for someone to tell you how the world works."
Join us as we hear from Andrew Barto and Richard Sutton, the 2024 #ACMTuringAward recipients as they discuss their work on #ReinforcementLearning.
This Python class offers a multiprocessing-powered Pool for efficiently collecting and managing experience replay data in reinforcement learning.
https://github.com/NoteDance/Pool
Discussions: https://discu.eu/q/https://github.com/NoteDance/Pool
AI Progress: Understanding the Myths Behind the Intelligence Explosion
As the AI landscape evolves, the concept of an 'intelligence explosion' raises questions about the future of AI development. This article delves into the current state of AI, addressing misconceptions...
https://news.lavx.hu/article/ai-progress-understanding-the-myths-behind-the-intelligence-explosion
Training AI to Persuade? - Jeremie & Edouard Harris on JRE
[AGI discussion, DeepMind] Welcome to the Era of Experience
https://storage.googleapis.com/deepmind-media/Era-of-Experience%20/The%20Era%20of%20Experience%20Paper.pdf
https://old.reddit.com/r/MachineLearning/comments/1k4zr1i/r_deepmind_welcome_to_the_era_of_experience
* threshold of new era in AI that promises unprecedented level of ability
* new generation of agents will acquire superhuman capabilities, learning predominantly f. experience
* paradigm shift, accompanied by algorithmic advancements in RL, will unlock new supra-human capabilities
Happy birthday to Cognitive Design for Artificial Minds (https://lnkd.in/gZtzwDn3) that was released 4 years ago!
Since then its ideas have been presented and discussed widely in the research fields of AI/Cognitive Science/Robotics and - nowadays - both the possibilities and the limitations of: #LLMs, #GenerativeAI and #ReinforcementLearning (already envisioned and discussed in the book) have become a common topic of research interests in the AI community and beyond.
Similarly also the topic concerning the evaluation - in human-like and human-level terms - of the current AI systems has become a critical theme related to the problem Anthropomorphic interpretation of AI output (see e.g. https://lnkd.in/dVi9Qf_k ).
Book reviews have been published on ACM Computing Reviews (2021) https://lnkd.in/dWQpJdkV and on Argumenta (2023): https://lnkd.in/derH3VKN
I have been invited to present the content of the book in over 20 official scientific events in international conferences, Ph.D Schools in US, China, Japan, Finland, Germany, Sweden, France, Brazil, Poland, Austria and, of course, Italy.
A news I am happy to share is that Routledge/Taylor & Francis contacted me few weeks ago for a second edition! Stay tuned!
The #book is available in many webstores:
- Routledge: https://lnkd.in/dPrC26p
- Taylor & Francis: https://lnkd.in/dprVF2w
- Amazon: https://lnkd.in/dC8rEzPi
@academicchatter @cognition
#AI #minimalcognitivegrid #CognitiveAI #cognitivescience #cognitivesystems
Implemented 18 RL Algorithms in a Simpler Way
https://github.com/FareedKhan-dev/all-rl-algorithms
Discussions: https://discu.eu/q/https://github.com/FareedKhan-dev/all-rl-algorithms
The article provides good insights into industry leaders such as Waymo, DeepMind, and Amazon demonstrate the transformative power of Reinforcement Learning (RL).
Takeaways:
RL drives autonomy and innovation across industries, but challenges like interpretability remain pivotal. Hybrid systems that blend RL and symbolic reasoning hint at breakthroughs in high-level decision-making.
Check out QwQ-32B! 
A 32B parameter model that rivals DeepSeek-R1 in performance, thanks to Reinforcement Learning. It excels in math, coding & general problem-solving. Open-weight on Hugging Face & ModelScope! Explore the future of AGI! 
https://qwenlm.github.io/blog/qwq-32b/ #AI #ReinforcementLearning #MachineLearning #Qwen #OpenSource
My colleagues at TU Delft are seeking to hire a postdoc to work on Applied Planning and Scheduling under Uncertainty, with applications in modelling supply chain scenarios for offshore wind farm installation: https://careers.tudelft.nl/job/Delft-Postdoc-in-Applied-Planning-and-Scheduling-under-Uncertainty-2628-CD/814890902/
How to formulate exploration-exploitation trade-off better than all the hacks on top of Bellman equation?
We can first of all simply estimate the advantage of exploration by Monte-Carlo in a swarm setting: Pitting fully exploitative agents against fully exploitative agents which have the benefit of recent exploration. This can be easily done by lagging policy models.
Of course the advantage of exploration needs to be divided by the cost of exploration, which is linear to the number of agents used in the swarm to explore at a particular state.
Note that the advantage of exploration depends on the state of the agent, so we might want to define an explorative critic to estimate this.
What's beautiful in this formulation is that we can incorporate autoregressive #WorldModels naturally, as the exploitative agents only learn from rewards, but the explorative agents choose their actions in a way which maximizes the improvement of the auto-regressive World Model.
It brings these two concepts together as sides of the same coin.
Exploitation is reward-guided action, exploration is auto-regressive state transition model improvement guided action.
Balancing the two is a swarm dynamic which encourages branching where exploration has an expected value in reward terms. This can be estimated by computing the advantage of exploitative agents utilizing recent exploration versus agents which do not, and returning this advantage to the points of divergence between the two.
DeepSeek R1: All you need to know 
The article covers various aspects of the model, from its architecture to training methodologies and practical applications. The explanations are mostly clear and detailed, making complex concepts like Mixture of Experts (#MoE) and reinforcement learning easy to understand.
New instance, new #introduction!
I'm a #DataScientist with a background in #ReinforcementLearning and #ElectricalEngineering. Well, that's what my resume says, but really I'm a #poet and a SF/F #writer. I love to play #DnD and other #TTRPGs.
I use they/them pronouns, and "Dr." not "Mr.", please and thank you.
I maintain a blog at www.seanpatrick.phd which includes a current list of publications, including my debut sonnet collection, "Love, Death, and Other Surprises."
