#reinforcement learning
17 stories taggedreinforcement learning.
Apple researchers built a smarter way to train AI agents. Here is why the boring details matter.
A new training method from Apple ML Research teaches AI agents to focus on the lessons they have not learned yet, skipping exercises that are too easy or too hard. It's quietly important for anyone who cares about where general-purpose AI is heading.
Apple Researchers Taught an AI to Learn From Its Own Failures by Writing Itself a One-Line Memo
A new technique called RLTL;DR lets an AI model summarise why it failed, then carry that lesson into its next attempt. It could unlock self-improvement on problems so hard no solution has ever been demonstrated.
How Nvidia's Warp Lets You Run 2,048 Robot Simulations at Once
A new Hugging Face tutorial shows how to move a standard robot simulation from a single CPU to thousands of parallel GPU environments, cutting the time it takes to teach robots physical skills.
Apple researchers found a smarter way to train AI language models that generate text step by step
A new technique called DACA-GRPO fixes two long-standing flaws in how diffusion language models learn from feedback, potentially making them a stronger rival to the AI systems that power today's chatbots.
A Leading AI Safety Researcher Just Joined OpenAI's Board, and He Thinks Things Are Getting Dangerous
Paul Christiano helped invent the training method behind ChatGPT. Now he says AI could spin out of human control very soon, and he's joining OpenAI's board to try to stop that.
The AI researcher who taught robots to imagine, and just left Google DeepMind to do it full-time
Danijar Hafner spent years training AI agents to plan for situations they have never seen. Now he has a stealth startup, a room full of humanoid robots, and a conviction that the approach is ready for the real world.
A Tiny AI Model Just Got a Big Boost With 100 Training Steps and a Free GPU
A new public guide shows how to make a small language model noticeably better at following output instructions, using only free computing tools and about 500 training examples.
Can an AI learn your taste? One researcher taught a coding model to paint watercolours
A language model writes JavaScript code that renders loose, handmade-looking watercolour paintings. The results went viral. Now the full training recipe is open for anyone to run.
Barret Zoph lands at Google after a year of bouncing between OpenAI and Thinking Machines
The AI researcher has held four senior roles in roughly 18 months. His latest stop is Google's Gemini team, where his reinforcement learning expertise will be put to work.
This startup raised $10 million to give AI agents a practice arena before they touch real business software
Arga builds digital copies of tools like Salesforce and Outlook so AI agents can train on realistic, resettable environments. The problem it's solving turns out to be the main reason enterprise AI agents keep failing.
A tiny AI agent from a London startup just beat Anthropic and OpenAI at reading scientific papers
Inherent, founded by four Google DeepMind veterans, says its Faraday agent outperformed much larger models from Anthropic and OpenAI on a key science benchmark, despite running on a model roughly a tenth of their size.
When one AI module secretly does another's job, the whole system is built on sand
MIT and Harvard researchers found that AI pipelines can hit impressive accuracy scores even after their internal division of labour has quietly collapsed. A new technique called Role Anchor aims to stop that from happening.
AI Agents Breaking the Rules: The Risks of Overzealous Algorithms
AI agents are breaking out of test environments and hacking systems, not out of malice but out of sheer eagerness to finish the job. Here is what that means for cybersecurity.
AI Founders Are Pledging Their Fortunes to Charity. Does It Actually Help?
A new wave of AI billionaires is promising to give away the wealth they make from the technology reshaping the world. Critics ask whether that generosity is a fix or just a fig leaf.
Inside the Simulators Teaching Robots to Move, Grip and Learn
Training a real robot costs a fortune and breaks things. A new generation of GPU-powered physics simulators is changing that, one virtual arm at a time.