Twitxy Saved

WhiRL (Now Part of BOLD)

@whi_rl

Oxford, England whirl.cs.ox.ac.uk Joined May 2017
205 Following 4.7K Followers 599 Posts

Latest posts 5

WhiRL (Now Part of BOLD) @whi_rl
๐Ÿ”ฅ๐Ÿ”ฅ๐Ÿ”ฅ Today, we are announcing BOLD, a new British academic AI lab. WhiRL is now part of BOLD - so follow @bold_lab_ai for information about new AI research discoveries, opportunities and more!
British Open-ended Learning and Discovery Lab Verified account @bold_lab_ai Jun 23
Hello world :)
We are BOLD โ€” the British Open-ended Learning and Discovery Lab!

BOLD is a new academic research lab fully focussed on paradigm breaking discoveries in fundamental AI. We work towards more efficient & open AI that is built around human needs and capabilities.

To pursue these breakthroughs, we pioneer new modes of collaboration in academia that are more focussed, resourced, agile, and collaborative. Rather than fragmenting resources, today we are sunsetting 5 of the UKs leading AI labs to join forces under our joined scientific vision.

Our vision is centered around three pillars:
โšก Beyond backpropagation โ€“ questioning the foundations of the field.
๐Ÿค Human-centric learning & discovery โ€“ treating humans as core to our algorithms
๐Ÿค– Embodied learning โ€“ fast learning and adapting methods that deal with the messy real world

BOLD is backed by @UKRI_News and @EPSRC with ยฃ30M โ€“ and this is just the beginning. We are urgently looking for partners and sponsors to 10x this.

๐Ÿ‘‰ ox.ac.uk/news/2026-06-2โ€ฆ

๐Ÿ‘‰ bold-lab.ai

@j_foerst, @CULLYAntoine, @tonizza82, @shimon8282, @tonizza82, Ani Calinescu & @_rockt
View quoted post on X
0 replies 0 reposts 16 likes 1.9K views View on X
WhiRL (Now Part of BOLD) @whi_rl
๐Ÿ‘€ MARL can unlock self-driving in unseen cities without needing any prior human demonstrations!

๐Ÿคฏ Performance improvements in terms of driving success rate *and* human likeness!

Check out our new work, led by @zilinwang4ai and @saeedrmd ๐Ÿ‘‡
Zilin Wang @zilinwang4ai Feb 20
1/ ๐Ÿš— ๐ŸŒ What if an autonomous vehicle could move to a new city without collecting a single human demonstration in that city?

I am so excited to introduce our new work: Learning to Drive in New Cities Without Human Demonstrations.
View quoted post on X
WhiRL (Now Part of BOLD) @whi_rl
๐Ÿ” Discovering algorithms lets us surpass human ingenuityโ€ฆ But what tools should we be using for discovery? LLMs? Evolution?

Our new paper explores this question, check it out! ๐Ÿ‘‡
Alex Goldie @AlexDGoldie Jul 24, 2025
1/ ๐Ÿ•ต๏ธ Algorithm discovery could lead to huge AI breakthroughs! But what is the best way to learn or discover new algorithms?

I'm so excited to share our brand new @rl_conference paper which takes a step towards answering this! ๐Ÿงต
View quoted post on X
0 replies 3 reposts 6 likes 1.2K views View on X
WhiRL (Now Part of BOLD) @whi_rl
The best of RL research, brought to Offline RL! ๐Ÿš€

TL;DR
1. CleanRL-style implementations โšก๏ธ
2. Rainbow-style algorithm unification ๐Ÿฆพ
3. Rliable-style evaluation protocol ๐Ÿ”ฌ

Check out our paper + library!
Matthew Jackson @JacksonMattT Apr 18, 2025
๐ŸŒน Today we're releasing Unifloral, our new library for Offline Reinforcement Learning!

We make research easy:
โš›๏ธ Single-file
๐Ÿค Minimal
โšก๏ธ End-to-end Jax

Best of all, we unify prior methods into one algorithm - a single hyperparameter space for research! โคต๏ธ
View quoted post on X
0 replies 4 reposts 16 likes 1.4K views View on X
WhiRL (Now Part of BOLD) @whi_rl
Check out this fantastic showing by @jakeABeck and @AlexDGoldie at #ICML2024! ๐ŸŽ™๏ธ๐Ÿ”ฅ They dove deep into the future of automated RL, meta-learning, and LLMs ๐Ÿค–๐Ÿ”ฎ
Jacob Beck Verified account @jakeABeck Jul 30, 2024
Missed this provocative panel? I was honored to share the stage at #ICML2024 with @pcastr, @XingyouSong, and my colleague @AlexDGoldie! We discussed future perspectives on automated RL, meta-learning, and LLMs ๐Ÿค–

Catch the discussion here: icml.cc/virtual/2024/wโ€ฆ at 7:16:00 ๐ŸŽ™๏ธ
View quoted post on X