Scaling Up Reinforcement Learning for Traffic Smoothing: A 100-AV Highway Deployment

AI learns to sync sight and sound

Imagine watching a video where someone slams a door, and the AI behind the scenes instantly connects the exact moment of that sound with the visual of the door closing – without ever being told what a door is. This is the future researchers at MIT and international collaborators are building, thanks to a breakthrough … Read More

Perplexity Introduces Comet—An AI-First Alternative to Traditional Browsers

Perplexity, a company already recognized for redefining how users interact with information through AI-powered search, has announced the launch of Comet, an ambitious AI-native web browser. Designed with an AI-first architecture, Comet is set to revolutionize how users navigate, retrieve, and interact with web content by combining browsing with real-time contextual assistance, summarization, and intelligent … Read More

A 100-AV Highway Deployment – The Berkeley Artificial Intelligence Research Blog

Training Diffusion Models with Reinforcement Learning

We deployed 100 reinforcement learning (RL)-controlled cars into rush-hour highway traffic to smooth congestion and reduce fuel consumption for everyone. Our goal is to tackle “stop-and-go” waves, those frustrating slowdowns and speedups that usually have no clear cause but lead to congestion and significant energy waste. To train efficient flow-smoothing controllers, we built fast, data-driven simulations that RL agents interact with, learning to maximize energy efficiency while maintaining throughput and operating safely around human drivers.

Overall, a small proportion of well-controlled autonomous vehicles (AVs) is enough to significantly improve traffic flow and fuel efficiency for all drivers on the road. Moreover, the trained controllers are designed to be deployable on most modern vehicles, operating in a decentralized manner and relying on standard radar sensors. In our latest paper, we explore the challenges of deploying RL controllers on a large-scale, from simulation to the field, during this 100-car experiment.

Study could lead to LLMs that are better at complex reasoning | MIT News

For all their impressive capabilities, large language models (LLMs) often fall short when given challenging new tasks that require complex reasoning skills. While an accounting firm’s LLM might excel at summarizing financial reports, that same model could fail unexpectedly if tasked with predicting market trends or identifying fraudulent transactions. To make LLMs more adaptable, MIT … Read More

Free Local RAG Scraper for Custom GPTs and Assistants • AI Blog

This web scraper runs entirely in your browser and is perfect for creating training data for AI models. It works by reading the website’s sitemap.xml file, making it particularly well-suited for modern platforms like Squarespace and Shopify that automatically generate sitemaps. The scraper preserves the structure of your content, including headings, paragraphs, lists, and tables, … Read More

Why AI-Driven Logistics and Supply Chains Need Resilient, Always-On Networks

Modern supply chains are extremely complex, intricate, and expansive, comprising many parties (like brokers, shippers, and warehouses) that must communicate and operate in a timely and organized manner. Like any ecosystem, one small disruption can affect the larger environment in unexpected and ruinous ways. Consequently, many enterprises have incorporated artificial intelligence (AI)-powered systems and applications … Read More

Real-life transformer: Drone morphs mid-air

Engineers at Caltech have successfully created a real-life “Transformer” – a groundbreaking development in robotics for aerial and ground operations. Named ATMO (short for Aerially Transforming Morphobot), the robot can transition from a flying drone to a rolling rover while still in mid-air. This innovative design offers a solution to a long-standing challenge faced by … Read More

Google AI Open-Sourced MedGemma 27B and MedSigLIP for Scalable Multimodal Medical Reasoning

In a strategic move to advance open-source development in medical AI, Google DeepMind and Google Research have introduced two new models under the MedGemma umbrella: MedGemma 27B Multimodal, a large-scale vision-language foundation model, and MedSigLIP, a lightweight medical image-text encoder. These additions represent the most capable open-weight models released to date within the Health AI … Read More

Repurposing Protein Folding Models for Generation with Latent Diffusion – The Berkeley Artificial Intelligence Research Blog

PLAID is a multimodal generative model that simultaneously generates protein 1D sequence and 3D structure, by learning the latent space of protein folding models.

The awarding of the 2024 Nobel Prize to AlphaFold2 marks an important moment of recognition for the of AI role in biology. What comes next after protein folding?

In PLAID, we develop a method that learns to sample from the latent space of protein folding models to generate new proteins. It can accept compositional function and organism prompts, and can be trained on sequence databases, which are 2-4 orders of magnitude larger than structure databases. Unlike many previous protein structure generative models, PLAID addresses the multimodal co-generation problem setting: simultaneously generating both discrete sequence and continuous all-atom structural coordinates.

AI shapes autonomous underwater “gliders” | MIT News

Marine scientists have long marveled at how animals like fish and seals swim so efficiently despite having different shapes. Their bodies are optimized for efficient, hydrodynamic aquatic navigation so they can exert minimal energy when traveling long distances. Autonomous vehicles can drift through the ocean in a similar way, collecting data about vast underwater environments. … Read More

Categories