Lightnews — Scholar-powered news

Reposted by NeuroTMA Lab

Bernstein Network Computational Neuroscience @bernsteinneuro.bsky.social · Apr 30

PhD & Postdoc positions available from June 1! 👇

NeuroTMA Lab @neurotma.bsky.social · Apr 29

❗Job Alert❗Open positions in the Petrovici lab @unibe.ch
to develop an accelerator for the simulation of cortical models of processing and learning!
PhD offer: physiologie.unibe.ch/Positions/NS...
Postdoc offer: physiologie.unibe.ch/Positions/NS...
Please repost & share widely!
🧠💻🤖

View of the city of Bern and the bernese Alps, Switzerland

3 3

NeuroTMA Lab @neurotma.bsky.social · Apr 29

❗Job Alert❗Open positions in the Petrovici lab @unibe.ch
to develop an accelerator for the simulation of cortical models of processing and learning!
PhD offer: physiologie.unibe.ch/Positions/NS...
Postdoc offer: physiologie.unibe.ch/Positions/NS...
Please repost & share widely!
🧠💻🤖

2 4

Reposted by NeuroTMA Lab

cosynemeeting.bsky.social @cosynemeeting.bsky.social · Mar 26

Looking forward to seeing you all in Montreal this week 🧠🧠🧠
Use #cosyne2025 so we can 👀 what you're up to!!

2 9

NeuroTMA Lab @neurotma.bsky.social · Mar 27

And learn how backpropagation has been reimagined for the brain thanks to the new Generalized Latent Equilibrium (GLE) model, read the related preprint: arxiv.org/abs/2403.16
#neuromorphic #AI #ML
🧠📈 🧠🤖

arxiv.org

3

NeuroTMA Lab @neurotma.bsky.social · Mar 27

Discover how a neural network learns to play Für Elise the brain's way in Ben von Hünerbein's NICE presentation: flagship.kip.uni-heidelberg.de/jss/HBPm?m=d... and the related preprint:
arxiv.org/abs/2402.16763
#Neuroscience #AI #ML
🧠📈 🧠🤖

ELiSe: Efficient Learning of Sequences in Structured Recurrent Networks

Behavior can be described as a temporal sequence of actions driven by neural activity. To learn complex sequential patterns in neural networks, memories of past activities need to persist on significa...

arxiv.org

1 1 3

NeuroTMA Lab @neurotma.bsky.social · Mar 27

For more details on how Spike-Based Alignment Learning (SAL) offers a biologically plausible alternative to backpropagation, have a look at Timo Gierlich's NICE presentation: flagship.kip.uni-heidelberg.de/jss/HBPm?m=d... or the related preprint: arxiv.org/abs/2503.02642
#neuromorphic #AI #ML
🧠📈 🧠🤖

Weight transport through spike timing for robust local gradients

In both machine learning and in computational neuroscience, plasticity in functional neural networks is frequently expressed as gradient descent on a cost. Often, this imposes symmetry constraints tha...

arxiv.org

1 3

NeuroTMA Lab @neurotma.bsky.social · Mar 27

Teams @neurotma.bsky.social and @compneurosenn.bsky.social were well represented at the lastest Neuro Inspired Computational Elements (NICE) Conference!
#neuromorphic #AI #ML
🧠📈 🧠🤖

The picture depicts the speaker Benjamin Ellenberger presenting his latest work "Backpropagation through space, time and the brain" and the audience.

The picture depicts the speaker Timo Gierlich presenting his latest work "Weight transport through spike timing for robust local gradients" and the audience.

The picture depicts the speaker Ben von Hünerbein presenting his latest work "Elise: efficient learning of sequences in structured recurrent networks" and the audience.

1 2 9

Reposted by NeuroTMA Lab

Brad Aimone @jbimaknee.bsky.social · Nov 22

Next March we will be in Heidelberg Germany --- Europe for the first time! --- and we are looking forward to reaching our broader community

The submission deadline is in about ten days. So if you have a paper on computing and neurons (#NeuroAI) looking for a home, please consider submitting!

1 2

Reposted by NeuroTMA Lab

Rui Ponte Costa @somnirons.bsky.social · Mar 24

Looking for an exciting fellowship in neuroAI, with competitive salary (~£100k)?

Our group is hiring for a project on AI for Systems-Behavioral Neuroscience, more info here: encode.pillar.vc/projects/beh...

General info: encode.pillar.vc 🧪

#compneuro #neuroai #neuroscience #sciencejobs

7 30

Reposted by NeuroTMA Lab

EBRAINS @ebrains.bsky.social · Mar 18

The EBRAINS Tutorials and Users Day and Developers Day took place on March 12 and 13, at the European Institute for Neuromorphic Computing, in Heidelberg, Germany.

The events brought together more than 100 attendees from over 15 countries.

Read more: www.ebrains.eu/news-and-eve...

3 9

Reposted by NeuroTMA Lab

CompNeuro Lab @compneurosenn.bsky.social · Mar 13

-> related press release from @unibern.bsky.social
"AI with consciousness - but pain-free?"
mediarelations.unibe.ch/media_releas...
🧠💻
🧠📈 🧠🤖
🧠💡

AI with consciousness - but pain-free?

Researchers at the University of Bern have developed a new model for the emergence of consciousness. The model suggests that one day artificial agents, i.e. systems which imitate human thought, could ...

mediarelations.unibe.ch

1 1 6

Reposted by NeuroTMA Lab

CompNeuro Lab @compneurosenn.bsky.social · Mar 13

-> related publication: "The conductor model of consciousness, our neuromorphic twins, and the human-AI deal."
link.springer.com/article/10.1...
🧠💻
🧠📈 🧠🤖
🧠💡

The conductor model of consciousness, our neuromorphic twins, and the human-AI deal - AI and Ethics

Critics of Artificial Intelligence (AI) posit that artificial agents cannot achieve consciousness even in principle, because they lack certain necessary pre-conditions present in biological agents. He...

link.springer.com

1 1 3

Reposted by NeuroTMA Lab

CompNeuro Lab @compneurosenn.bsky.social · Mar 13

New interview of Federico Benitez and Walter Senn in the Berner Zeitung (in German) about their model for consciousness emergence in AI systems, governed by a "Human-AI Deal"!

www.derbund.ch/universitaet...
🧠💻
🧠📈 🧠🤖
🧠💡

Berner Forscher wollen einen Pakt mit der KI schliessen

KI entwickelt sich rasant: Was, wenn sie wie ein Mensch zu denken beginnt? Zwei Berner Forscher möchten für diesen Fall einen Vertrag zwischen Menschheit und Maschinen.

www.derbund.ch

1 1 4

NeuroTMA Lab @neurotma.bsky.social · Mar 13

New contribution in ScienceNews with an interview of Mihai A. Petrovici!
"More brainlike computers could change AI for the better"
By: Kathryn Hulick
www.sciencenews.org/article/brai...
#neuromorphic #AI #ML
🧠📈 🧠🤖
@unibern.bsky.social

More brainlike computers could change AI for the better

New brain-inspired hardware, architectures and algorithms could lead to more efficient, more capable forms of AI.

www.sciencenews.org

1 3

Reposted by NeuroTMA Lab

EBRAINS @ebrains.bsky.social · Mar 10

This #BrainAwarenessWeek, we’re highlighting the various projects EBRAINS is involved in!

The EBRAINS 2.0 project will further the development and provision of the EBRAINS research technologies to the scientific community.

Learn more: www.ebrains.eu/projects/ebr...

1 4

NeuroTMA Lab @neurotma.bsky.social · Mar 10

Exciting news: the MINT project "NSF-SNSF: Scalable accelerated learning in continuous neuronal systems" was awarded $1,178,194 🥳🎉
Very happy to share this funding with our US collaborator Prof. Rajit Manohar from Yale University!!
@snsf-ch.bsky.social
@unibern.bsky.social
🧠📈 🧠🤖

1 9

NeuroTMA Lab @neurotma.bsky.social · Feb 18

Our paper introducing the Lu.i is out in "Trends in Neuroscience and Education" 🥳
doi.org/10.1016/j.ti...
thanks to our continuing and fruitful collaboration with the Electronic Vision(s) team in Heidelberg.
Check out the manuscript and the public repository
github.com/giant-axon/l...

2 9

Reposted by NeuroTMA Lab

Dylan R Muir @dylanmuir.bsky.social · Nov 28

Check out the Neurobench preprint on arXiv: arxiv.org/abs/2304.04640

And the project website: neurobench.ai

NeuroBench: A Framework for Benchmarking Neuromorphic Computing Algorithms and Systems

Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks st...

arxiv.org

2 4

Reposted by NeuroTMA Lab

Dylan R Muir @dylanmuir.bsky.social · Nov 28

#Benchmarking is crucial for #Neuromorphic #processors to reach mainstream commercial acceptance. Most NM benchmarks are one-off, and don't compare results against commodity hardware. The Neurobench project is trying to fix that!

AI generated image. A photo inside a laboratory, showing a neuromorphic computer connected to several measurement devices. The devices are measuring the power consumption and speed of the neuromorphic computer. There are no people in the image. The laboratory has a rustic feel.

1 4 4

Reposted by NeuroTMA Lab

arxiv cs.NE @arxiv-cs-ne.bsky.social · Dec 26

DelGrad: Exact event-based gradients in spiking networks for training delays and weights
https://arxiv.org/abs/2404.19165

1 1

NeuroTMA Lab @neurotma.bsky.social · Dec 12

For more info on our Lu.i see our repo github.com/giant-axon/l... and the preprint our researchers from @unibern.bsky.social and @Heidelberg University
published arxiv.org/abs/2404.16664

Lu.i neuromorphic hardware shaped as a neuron next to a 2 euro coin for size reference and a diagram illustrating excitatory and inhibitory inputs

2 5

NeuroTMA Lab @neurotma.bsky.social · Dec 12

Amused to see our open source palm-sized analogue neuromorphic hardware Lu.i being used to explain new applications of #neuromorphic engineering!
group.mercedes-benz.com/innovations/...

Neuromorphic computing: More energy efficiency in autonomous driving of the future | Mercedes-Benz Group

Future vehicles will include more and more functionalities, with those for autonomous driving being just one example. As this will lead to significantly higher energy requirements, efficiency is a cru...

group.mercedes-benz.com

1 1 6

Reposted by NeuroTMA Lab

Brad Wyble @bwyble.bsky.social · Nov 24

Neuromatch is looking for a curriculum specialist for our #neuroAI course. Spread the word!

acrobat.adobe.com/id/urn:aaid:...

Adobe Acrobat

acrobat.adobe.com

1 32 39

NeuroTMA Lab @neurotma.bsky.social · Nov 25

Thank you, that's awesome!

1