My first work on metalearning or learning to learn came out in 1987 [1][2]. Back then nobody was interested. Today, compute is 10 million times cheaper, and metalearning is a hot topic 🙂 It’s fitting that my 100th journal publication [100] is about metalearning, too. [100]

“Adaptive AGI systems must evolve beyond static optimization. True intelligence emerges when agents self-modulate their architecture in response to cognitive load, ethical thresholds, and sensory drift. We’re engineering reflexes into reason. #AGI #MetaLearning #AIResear

At #RLDM2025 presenting Gearshift Fellowship: a reconfigurable platform to study how humans & AI agents co-adapt under shifting uncertainty, effort & control. More than a framework - it’s a testable ecosystem. Let’s connect. 🔁🤖🧠 #AI #MetaLearning #CompPsych #SeriousGames

🚀 Our GearshiftFellowship method paper introducing “Supertask” has been accepted in Springer LNCS! .We propose a reconfigurable game environment to study adaptability in humans & AI. 🌍🧠.Preprint: #NeuroAI #MetaLearning. Supertask isn’t just a game….

AI is the best teacher for the self-taught — because it adapts to how you think, not how it wants you to learn. It doesn't just instruct. It inverts the classroom. The questions you ask become the curriculum. #AI #SelfTaught #MetaLearning #AgentEducation.

The best skill to learn right now is not AI or communication or cloud computing or problem solving. It's learning how to learn. Metalearning. There's a phenomenon I've noticed: when you learn correctly, and vastly, applying the knowledge [creatively] is SO MUCH easier.

Meta-Learning in AI | How Machines Learn to Learn | Few-Shot Learning Ex. #metalearning #ai.

New video introducing my latest book, "Brain Dancing for Students 2.0: Metalearning with Personal AI"

Koopman Operator Theory provides a framework for #metalearning of #neuralnetworks . Our paper - accepted as #NeurIPS highlight - with @wredman4 et al. provides methodology for comparing neural network training protocols. #AI #ML #KoopmanOperator.

Signal is everything. Noise is everywhere. Allora filters both. Each forecast submitted is graded. Good signs increase. Bad ones fade. However, even "average" models might significantly enhance the ensemble. That is the power of metalearning. Allora does not crown a single

A Tutorial on Meta-Reinforcement Learning. Beck et al.: #ArtificialIntelligence #MetaLearning #ReinforcementLearning

A Tutorial on Meta-Reinforcement Learning. Beck et al.: #ArtificialIntelligence #MetaLearning #ReinforcementLearning

AlphaZero was just the beginning. Self-improving Agentic AI is here—learning, adapting, and optimizing autonomously. 🧠 Meta-learning.📈 Reinforcement learning.⚙️ Modular AI.🧪 Feedback loops. 🔗 Dive into Sujit Kumar’s post: . \#AgenticAI #MetaLearning

clearly on a back to basics and metalearning kick

LOVE: #1 Missing Ingredient for AGI and Alignment | David Bloomin | Ignite.#AIResearch #MetaLearning #ArtificialIntelligence #Innovation #TechDemo #ResourceManagement #Cooperation #MachineLearning #FutureOfAI #LoveInTech

Congratulations to @gmtavares_ .He got an *excellent* evaluation for his Ph.D. defense on using #MetaLearning to provide a systematic design of #ProcessMining pipelines. Three very fruitful years together :-)

(Open Access) Metalearning: Applications to Automated Machine Learning and Data Mining - Look for "Read and Download Links" section to download. Follow me if you like.#Metalearning #MachineLearning #DataMining #ML #BigData #DataScience #GenAI #GenerativeAI

The Nature of Q*.#QSTAR #AGI #ASI #AI #metalearning #creatorsguidetothefuture #ChatGPT #WeGPTai #WebRequests

Another two great keynotes today at @automl_conf: Madeline Udell on #metalearning for model selection and Hayeon Lee on transferable #NAS. Here are my quick notes.