![Mirco Musolesi Profile](https://pbs.twimg.com/profile_images/1155577278946074626/vERyDAGJ_x96.jpg)
Mirco Musolesi
@mircomusolesi
Followers
3K
Following
5K
Statuses
5K
Professor of Computer Science. Machine Intelligence Lab. @AI_UCL, @UCLCS, @UCL. AI/ML, computers and books. https://t.co/Eh3ZcBJChM
London, United Kingdom
Joined October 2008
RT @royalsociety: The Royal Society's Newton International Fellowships are now open for applications. These fellowships are for non-UK earl…
0
21
0
Our paper on moral alignment for LLM agents has been accepted at #ICLR2025. Well done @liza_karmannaya! Preprint:
Excited to announce that I'll be presenting this work at ICLR @iclr_conf in Singapore later this year! Can't wait for the chats with fellow researchers working on alignment & agents 🤖🧠🚀
0
1
18
@GNRailUK @HenWelder Well, we are now entering Royston. At Letchworth, we might break the @GWR World Record for number of people in a train carriage. You should send a judge.
2
0
2
RT @SilverJacket: New research: AI writing is improving, but it still can’t match human creativity. My latest for @ScienceMagazine @Newsfr…
0
9
0
I like this idea of mixing LLMs and quaint communication modalities like making a phone call. They should really add a fax option.
You can now talk to ChatGPT by calling 1-800-ChatGPT (1-800-242-8478) in the U.S. or by sending a WhatsApp message to the same number—available everywhere ChatGPT is.
0
0
5
@GNRailUK So last week it took me three hours to go home _twice_ and now we start the week with zero trains to KGX. I was going to say that you should get nationalised now, but then I reminded myself that National Rail (running the tracks) is already nationalised. Commuters are powerless.
0
0
1
RT @RSocPublishing: 📢Are you an early career researcher, interested in how science is communicated, reported, & the evolution of the scient…
0
33
0
Our paper with @Gionceschelli "On the creativity of large language models" is now out in AI&Society as open access article.
0
0
5
Our paper with Alexei Pisacane and @VictorDarvariu "Reinforcement Learning Discovers Efficient Decentralized Graph Path Search Strategies" will be presented today at the @LogConference 2024 at 3pm UK time. #log2024 Link to the paper:
1
3
12
Our "manifesto paper" on the study of human behaviour and cognition in-the-lab & in-silico & in-the-wild (and back) co-authored with the students of the first cohort of our @LeverhulmeTrust Doctoral Training Programme on the Ecological Study of the Brain is finally out!
Out today in @royalsociety open science coauthored with our *first* Ecobrain students and supervisors. The paper presents our take on Real-World Cognitive Science/ Neuroscience. @UCLBrainScience @UCLPALS
0
1
4
The best reading while waiting for the results of the US election at 3 AM anyway.
New preprint on a solution for neural network pruning based on information theory is out! In particular, the paper presents an effective technique for pruning nodes, while maintaining at the same time the mutual information between adjacent layers.
0
0
1
New preprint on a solution for neural network pruning based on information theory is out! In particular, the paper presents an effective technique for pruning nodes, while maintaining at the same time the mutual information between adjacent layers.
I'm happy to announce a new pre-print! In this paper, @mircomusolesi, Steve Hailes, and I introduce MIPP: a pruning technique that preserves the mutual information between layers. Explore the work here:
0
4
6
RT @cecim: This mobile app study aims to collect data of respiratory tract infections progression through "respiratory" sounds. Please do p…
0
3
0