Behavioral cloning mystery
I've recently started working in robotics, where I've had the chance to train many behavioral cloning (BC) and reinforcement learning (RL) policies on real robot data. My impression so far is: real-world demonstration data is very different from the data in existing simulated RL benchmarks (e.g., D4RL, OGBench, etc.). In particular, I've noticed that there are many mysterious phenomena that appear with real robot data, which are not easily observed in many standard BC and RL benchmarks. These mysteries really confused me, so I decided to properly study them. Unfortunately, it's really difficult to investigate this in the real world, because nothing is fully reproducible in robotics. In the real world, results depend on everything, including lighting conditions, background, reset distribution, robot temperature, and so on. This means that even when you suspect that something weird is happening, you can't be 100% sure whether it's actually a thing or just noise. To do proper science, I b
I've recently started working in robotics, where I've had the chance to train many behavioral cloning (BC) and reinforcement learning (RL) policies on real robot data. My impression so far is: real-world demonstration data is very different from the data in existing simulated RL benchmarks (e.g., D4RL, OGBench, etc.). In particular, I've noticed that there are many mysterious phenomena that appear with real robot data, which are not easily observed in many standard BC and RL benchmarks. These mysteries really confused me, so I decided to properly study them. Unfortunately, it's really…
saved by
related reading
- State of Robot Learning, December 2025vedder.io
- [2507.09061] Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous Controlarxiv.org
- Sporks of AGIsergeylevine.substack.com
- Scalable Behavior Cloning with Open Data, Training, and Evaluationabc.bot
- Behavior Cloning is Miscalibrated — AI Alignment Forumalignmentforum.org
- How Claude Performs on Robotics Tasks \ Anthropicanthropic.com
- Supervised Policy Learning for Real Robotssupervised-robot-learning.github.io
- Ch. 21 - Imitation Learningunderactuated.mit.edu
- How Can We Make Robotics More like Generative Modeling? | Eric Jangevjang.com
- Many Small Steps for Robots, One Giant Leap for Mankindnotboring.co
- A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulationtoyotaresearchinstitute.github.io
- ABC-130K: The largest open source teleoperation datasetxdof.ai