Stefano Sarao Mannelli
@stefsmlab
Followers
684
Following
453
Media
23
Statuses
209
Assistant Professor @chalmersuniv and @goteborgsuni 🌱 ML + CogSci + StatPhys 🔎 @stefsm.bsky.social 🦋
Gothenburg, Sweden
Joined January 2010
Overdue announcement of our accepted @NeurIPSConf paper [ 🎉. Challenging the assumption that bias remains consistent throughout training, we show how data properties impact learning at different timescales, leading to surprising accuracy shifts!. 🧵 1/4
1
5
37
RT @EngelTatiana: Out today in @Nature: we show that individual neurons have diverse tuning to a decision variable computed by the entire p….
0
88
0
RT @Shoelim8: 📣 We are hiring! Want to move to beautiful Stockholm 🇸🇪 and work on cutting-edge ML research?.Join our group and help push th….
0
3
0
RT @smnlssn: We are looking for someone to join the group as a postdoc to help us with scaling implicit transfer operators. If you are inte….
0
16
0
RT @raillabwits: Great news! This work has also been accepted into RLC for the Journal to Conference Track. Congrats @geraudnt and teams fo….
0
9
0
🎙️= Maria Chiara Angelini, @giuseppe_durisi, @fgerace_ , Babak Hassibi, @Shoelim8, Bernhard Mehlig, Ayca Ozcelikkale, @rosalbapacelli, Mauro Pastore, @saglietti_luca, Johan Wästlund.⏱️= Devdatt Dubhashi, Ashkan Panahi, Me.
0
0
2
It is a great pleasure to host the Workshop in Advancements in High-Dimensional Methods for Machine Learning next week at @chalmersuniv!. We have a great set of speakers, and I look forward to the talks and discussions!. 🔗
1
4
17
RT @sebastiangoldt: Really happy to see this paper out, led by @nishpathead in collaboration with @stefsmlab and @SaxeLab: we apply the sta….
0
7
0
0
0
2
Our paper just came out in PRX! . Congrats to @nishpathead and the rest of the team. TL;DR : We analyse neural network learning through the lens of statistical physics, revealing distinct scaling regimes with sharp transitions. 🔗
The RL-Perceptron model grants new insights into the capabilities of policy learning in neural networks, including identifying trade-offs between #speed and #accuracy, and a transition between “easy” and “hybrid-hard” learning regimes. Learn more:
1
3
20
RT @summerfieldlab: AISI has published its research agenda. There is so much amazing work going on at AISI. Expe….
0
3
0
I can definitely confirm—my time at PCS was fantastic. I really enjoyed the experience, and several of my current colleagues and collaborators were fellow students in the program.
This is a fantastic master program. Students of the international track spend 1 semester in Trieste ( at @ictpnews and @Sissaschool ), 1 in Turin ( @PoliTOnews ), and 1 in Paris (Sorbonne / Saclay / Paris-Cité ) + final proj. Apply! Deadline May 25.
0
0
5
RT @SCSLWorkshop: 🚨 Join us at the Workshop on Spurious Correlation & Shortcut Learning (SCSL) at #ICLR2025!.@iclr_conf .🗓️ April 28, 2025….
0
12
0
The first poster is actually #460, but I am actually still queueing at the registration. First memorable experience of this ICLR is the longest queue of my life. 😅.
1
0
4
Also @iclr_conf and it is going to be a busy week!.1. Today at 10am, Optimal Protocols for Continual Learning via Statistical Physics and Control Theory #342.2. Tomorrow at 10am, A Theory of Initialisation's Impact on Specialisation #342.3. Speaking at @SCSLWorkshop o Monday.
1
2
14
RT @devonjarvi5: Our paper, “Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks” will b….
0
16
0
RT @smnlssn: Check out cool new work from our group in collaboration with Pfizer and AstraZeneca, lead by Julian Cremer and Ross Irwin on F….
0
13
0
Amazing news! Congrats everyone and especially @devonjarvi5!.
I'm excited to announce that I have received the Thuthuka Research Grant from the @NRF_News! I will use this grant to start my own lab focused on comp neuro and ML theory at @WitsUniversity: the @caandl_lab! This is a joint effort with @VictoriaWi83762, @stefsmlab and @geraudnt.
0
0
4