🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

trust fund wouldn't exist either – me and danni met on this course. grateful to Music at Brookes for a joint honours degree (english & music) that i increasingly realise was innovative and brave. i was accepted without reading a note of music & now i'm a music lecturer

Playing GPS with Tobit Flatscher with some cool lighting and cold! pc Lenny Freissmuth

Crazy times. Anyways, excited to unveil JaxMARL! JaxMARL provides popular Multi-Agent RL environments and algorithms in pure JAX, enabling an end-to-end training speed up of up to 12,500x! Co-led w/ Alex Rutherford Ben Ellis Matteo Gallici Post: blog.foersterlab.com/jaxmarl/

Beautiful research at Oxford Robotics Institute

Great news, robotics enthusiasts! 🤖 Due to popular demand, we're extending the deadline for the 5th UK Robot Manipulation Workshop poster submissions to December 8th, 2023! For full program, submissions and registration to the workshop visit robot-manipulation.uk

I'm at NeurIPS this week to present this paper! I'm always interested in chatting about RL/world models/planning 😀 I'm hoping to learn as much as possible this week about LLMs/foundation models for sequential decision-making 😍

New Article: "Right Place, Right Time: Proactive Multi-Robot Task Allocation Under Spatiotemporal Uncertainty" by Street, Lacerda, Mühlig, and Hawes jair.org/index.php/jair…

The final act from charlie street's thesis. It addresses the problem of proactively tasking robot teams to efficiently service tasks that have some probability of occurring at particular locations/time points. Wonderful work from Charlie as usual :)

We are looking to recruit a student for a PhD in AI for robotics, more specifically in decision-making for an autonomous mobile robot working in a nuclear decommissioning setting. UK funding available: robots.ox.ac.uk/~nickh/icase24/

My custom LEGO design of an ANYbotics ANYmal C (CERBERUS SubT’s version) is finally complete!

My group in Engineering Science, Oxford, Oxford Robotics Institute is looking for talented research students passionate about robot learning. Interested in doing a PhD researching efficient and versatile world models for robotics and beyond? This one may be for you…

I just passed my PhD defence Oxford Robotics Institute! Now back to building exo intern

Had a great time at #RSS2024 so far. I presented parts of our work on contact-rich control under partial observability in the dexterous manipulation workshop: openreview.net/pdf?id=7sKUwZv… Ping me if you want to chat more about our work and share ideas on contact-rich control!

Read latest #openaccess paper: Planning under uncertainty for #saferobotexploration using #Gaussianprocess prediction by A. ✨ Matt Budd Paul Duckworth Maurice Fallon Nick Hawes 🔓bit.ly/3MAWdXv Oxford Robotics Institute Oxford Dynamic Robot Systems

🧵1/ How should we choose levels for RL agents to train on? 👾 While regret is good theoretically, it is intractable to compute, so most methods use an approximation. We investigate these scoring functions and find that instead, using learnability (defined below) does better! 🎢

My incredible student Branton DeMoss has made some really amazing headway on understanding the complexity of networks as they learn, and applied this to grokking. Take a read!