Sim2Real YouTube video by Montreal Robotics

Link to the video is here: youtu.be/ItDjkBmiyc0

After a busy start to the year, I have a new lecture to share. Generalist models have been making good progress, but #sim2real just works. In this lecture, I explain domain randomization, simulation limitations, and connections to generalization.

The sub-agent teams are getting very good. So good, I may have to start providing my students with management textbooks soon.

@obsidian.md + Claude = : 🔥
Having your AI agent help organize and process your thoughts and ideas all on your computer is amazing.

Update on workshop dates are actually 9-14 of August.

Just a few days left to submit your work to @rl-conference.bsky.social ! All attendees will get an awesome show from Cirque du Soleil.

⏳ Application deadline: March 11th, 2026
🎯 Cohort size: Limited to 25 participants
👉 Apply now and join an immersive week of learning, collaboration, and innovation in Robotics & AI:
docs.google.com/forms/d/e/1F...

🎓 Audience: Undergraduate, Master’s, PhD students, and industry professionals
🧠 Prerequisites: Background in #ML, #AI, or #Robotics
🗣️ Language: English
💸 Cost: FREE (food, travel, and lodging covered)

🚀 Montreal #Robotics Summer School 2026 — Applications Now Open!
📅 Dates: August 2–7, 2026
📍 Location: @mila-quebec.bsky.social a (in person)
🤖 Program: 1 day of activities + 5 days of intensive Robotics & AI training (theory and hands-on)

Better lucky than good. :(

3 on 3 is maddness.

Omg, this game is nuts.

For more details about the challenges of BC, check out my blog neo-x.github.io/blog/2026/02....
Or go straight to the code to investigate yourself.
github.com/milarobotlea...
I will be releasing the homework assignments soon.

I am rerunning my class on robot learning this year, and I plan to push many code examples to help others get to the ugly details fast. One of these details is how BC gets off track as network sizes change. Blog and notebook below.

GitHub - milarobotlearningcourse/mini-grp Contribute to milarobotlearningcourse/mini-grp development by creating an account on GitHub.

The code is designed to be low-resource usage, so many can try it out. Everything can run in free Colab, except the last OpenVLA and pi0-type models.

Code: github.com/milarobotlea...

I have updated my tutorial on making Vision Language Action models. This tutorial starts with a basic Transformer and walks people through the steps to transform it into a full VLA that uses PaliGemma as the pretrained VLM. Links below.

World Modeling Workshop - Day 1 YouTube video by Mila - Institut québécois d'IA

Yes! Here are the streaming links for the first two days.

February 4: youtube.com/live/7Gyuar7...

February 5: youtube.com/live/GO-ct_v...

@mila-quebec.bsky.social

I am looking forward to our first round of speakers tomorrow at @Mila_Quebec #worldmodels workshop.
world-model-mila.github.io

Is the deadline really the day before the #ICRA2026 paper notification?

How, after billions of dollars spent on code generation tool development they still can't reliably generate a working Dockerfile...

I am finally fully benefiting from making my lecture content in LaTeX. Creating new content powered by LLMs to make examples and translate my content to a webpage and a book is a breeze. Just need to figure out how to add references faster.

Reinforcement Learning Conference (RLC) was added to the AI conference DL countdown.

rl-conference.cc
March 1: Abstract DL (AoE)
March 5: Submission DL (AoE)

Conference: Montreal, Quebec, Canada,
August 16th -19th, 2026.

I can get behind this reasoning. Mistakes like these are sign other care was skipped. Similar to double checking notations, grammar, structure, that if not done well make a paper unclear and will contribute to non reproducibility.

Maybe if we work collectively we can distribute the work across the teachers. I am planning on making leader boards for the assignments in my class. I almost have it working and can share. Code generation has saved me a lot of time.

Another exciting year for more RL! Submit your work to the RL conference and join us to talk a out how to make RL even better.

It is refreshing.

I'll be giving a talk tomorrow in the Embodied World Models for Decision Making workshop at #NeurIPS. I will present a number of recent works from my lab on combining foundational models with planning and RL.
When: 11:30-noon
Where: Level Room 30A-E

For those interest in my lab checkout the labs research statement.
fracturedplane.com/assets/proje...

GitHub - milarobotlearningcourse/micro_vlm Contribute to milarobotlearningcourse/micro_vlm development by creating an account on GitHub.

Here is the link to the code: github.com/milarobotlea...
Here is the link to the tutorial: youtu.be/AVialKJVDfE