Guillaume Bellec
@BellecGuill
Followers
956
Following
2K
Media
37
Statuses
419
Postdoc researcher interested in neuroscience and machine learning. Currently working at EPFL with Wulfram Gerstner.
Graz, Austria
Joined February 2018
RT @ClaireWyart: HIRING🧿 Excited about molecular biology & genetics 🧬🧬🧬? Wanna live in Paris? Join our team @wyartlab as a lab manager ! Pe….
0
10
0
RT @JeanRemiKing: Our latest brain-to-image decoding model is now available on 🤗 HuggingFace: . "Dynadiff: Single-stage Decoding of Image….
0
51
0
RT @lostanlen: Our team is hiring a postdoc in Audio AI! . What: speech, music, bioacoustics.How: multiresolution neural networks in the ra….
0
18
0
RT @wellingmax: I just want to point out how unbelievably offensive it is that the VP of the US (and Musk) courts with a neo-nazi party in….
theguardian.com
US vice-president meets with AfD leader Alice Weidel as Trump administration courts European populist parties
0
45
0
RT @chethan: How can we build BCIs that operate robustly in the face of nonstationarities?. Our @NeurIPSConf Datasets and Benchmarks paper….
0
19
0
RT @bsimsek13: 📢 I'm on the faculty job market this year! . My research explores the foundations of deep learning and analyzes learning and….
0
22
0
RT @jmourabarbosa: We did it. We killed neuro twitter and it's resuscitation it's happening somewhere else. Come join us!. .
thetransmitter.org
Daily neuroscience-related posts on the social-media platform this week have increased more than 400 percent, on average, compared with October.
0
37
0
Discussions are becoming really cool under the blue sky! đź’™ . Science is more fun on the other side. See u there.
0
1
3
RT @dpaleka: It has not been reported much, but I believe ETH Zurich has, as of last week, banned new Master and PhD students who attended….
0
75
0
RT @ShahabBakht: This is a great opportunity to secure funding for postdoc. My lab has a few research directions that align with the IVAD….
0
7
0
RT @NeuroPSI_saclay: We grieve the passing of Dr. Yves Frégnac. Yves has shaped the study of cortical processing and plasticity. He herald….
0
40
0
RT @franceinter: "Nous n'avons plus de voix sur la scène internationale", s'indigne Dominique de @Villepin, évoquant Gaza "dont plus person….
0
4K
0
RT @schmutz_val: 1. Synaptic weight scaling in O(1/N) self-induces a form of (implicit) spatial structure in networks of spiking neurons, a….
arxiv.org
The dynamics of spatially-structured networks of $N$ interacting stochastic neurons can be described by deterministic population equations in the mean-field limit. While this is known, a general...
0
7
0
RT @ylecun: Many (rich) people on the Right do not actually believe in democracy. They have a huge superiority complex. They think the popu….
0
550
0
RT @francoisfleuret: It's bizarre that a pencil and a piece of paper are such a powerful brain extension. This is not so many bits compared….
0
10
0
RT @wonderofscience: A passengers close-up view of an engine failure on a Boeing 777 in 2021. đź“˝: Chad Schnell .
0
265
0
RT @Tantacrul: Oh, for the love of God. Right, if anyone is ever considering being a patron for me or any other creator, please do not do….
0
940
0
This changes the perspective for me:. (1) ML at scale with event-based network is not impossible. (2) But it needs more careful theory! Even when you think you control everything (equivalent with ReLU), gradients can be exploding with the wrong parametrization of the model.
0
0
1
Many thanks to everyone involved from EPFL @LCN_EPFL and IBM @IBMResearch . I love this paper -- because training with spike-timing was exciting in the 90s and early 2000s, and was then considered impossible and incompatible with ML after the emergence of deep learning.
1
0
1
This enables a workflow to build single-spike CNN with ~0.3 spk per neuron for spiking hardware with unprecedented accuracy:. (1) Convert a pre-trained ReLU CNN to a B1-model . (2) Fine-tune with the hardware in the loop to account for analog/quantized numerical imprecision. 7/7
1
0
1