Termsslug: imitation-learning

Imitation Learning

模仿学习

A method where agents learn strategies by observing expert behavior

Curated Intelligence

13 items · Page 1/2
MIT Technology ReviewT2·Industry
58 pts
Tau Robotics Deploys Teleoperated Humanoid Cleaners to Harvest Training Data

San Francisco startup Tau Robotics is offering a humanoid-robot cleaning service in which operators teleoperate the machines via head- and wrist-mounted cameras to wipe counters and fold laundry, while harvesting real-home data to train future autonomous models. The company prices cleanings below human rates to attract data-generating jobs and has completed only a few hundred sessions, underscoring how much human assistance home robots still require.

Rohan Paul (@rohanpaul_ai)T2·Models
78 pts
Verne Robotics Open-Sources V0.5 World Action Model with Stronger Robustness

Verne Robotics released V0.5, an open-weight world action model built on the Wan2.2 video generator and trained on over 1.7M demonstrations, predicting the next second of arm, gripper and mobile-base motion. It learns new industrial tasks from under two hours of demos and degrades far less than Physical Intelligence's π0.5 under clutter and lighting shifts.

Rohan Paul (@rohanpaul_ai)T2·Agents
88 pts
LiberAI Releases Liber-0: Glove Data Replaces Robot Demos

Chinese startup LiberAI released Liber-0, a robot foundation model trained on data from lightweight sensor gloves. It achieved a 36% success rate on new tasks without any robot demonstrations. Experiments showed that 10K hours of glove motion data outperformed 2M hours of internet video, significantly lowering the barrier for embodied AI data acquisition.

IT之家 · 智能时代T2·Agents
68 pts
Tesla Builds 'Optimus Academy' with 10k Robots for Autonomous Training

Tesla is establishing an 'Optimus Academy' to train humanoid robots by deploying 10,000-20,000 units for autonomous practice and data accumulation. Elon Musk likens the training process to raising a child, noting that unlike autonomous driving, Optimus cannot rely on fleet data and must learn through observation and hands-on practice. Production has begun at the Fremont factory, with customer deliveries expected by late 2027.

Rohan Paul (@rohanpaul_ai)T2·Agents
78 pts
Maxinsights Raises $30M Seed to Launch MaxGlove for Robot Force Data Collection

Embodied AI data firm Maxinsights secured a $30M seed round and launched MaxGlove, a device capturing tactile and force data from human hands in real homes. This addresses the critical gap in robot manipulation training where vision-only data fails to capture force dynamics, aiming to build a million-hour dataset of high-quality operational signals.

MIT Technology ReviewT2·Agents
42 pts
Purdue Team Develops AI Robotic System for Stroke Rehabilitation

A Purdue University team developed a robotic rehabilitation system combining generative AI and real-time force feedback to mimic human therapists and provide precise assistance to stroke patients. Currently in initial experimental stages, the system aims to extend rather than replace therapist capabilities, with plans for long-term clinical studies and potential industrial applications.

IT之家 · 智能时代T2·Agents
68 pts
HUMXN Offers Free Home Repairs to Collect Data for AI Robots

AI lab HUMXN launches a 'free home services for data' model, subsidizing plumbing and electrical repairs to record skilled workers' real-world tasks for training embodied AI robots. This approach addresses the limitation of internet text and images in capturing physical interaction details, initially rolling out in Minneapolis, Chicago, and Miami to explore compliant acquisition of high-quality real-scenario training data.

Hugging Face BlogT1·Agents
48 pts
Amazon Demos Strands Agents and LeRobot Data Loop

Amazon and Hugging Face demonstrate a closed-loop workflow for robotics using Strands Agents, LeRobot, and Storage Buckets. The system enables agents to record demonstrations, train policies directly from the Hub, and deploy them back to hardware, maintaining a consistent LeRobot data format throughout.