I am thrilled to present tomorrow my #NeurIPS paper on #GradientFlossing, which is a novel approach to tackling the exploding/vanishing gradient problem based on concepts from dynamical systems theory (Lyapunov exponents):
https://t.co/fgQrCwh7GN
1/n
So happy to see this finally out! 🥳Published today in @NeuroCellPress
https://t.co/CZwYXCUaEU
@Jastyn_Poepplau spearheaded the study of prefrontal circuits during late development and identified a “critical period” of cognitive processing. Great teamwork!👏
Short thread below👇
Freely access electrophysiological, morphological, and transcriptomic data measured from thousands of individual mouse & human brain cells via the Allen Cell Types Database: https://t.co/hgwCMQGeNU
#openscience#SfN23
How to analyze comput. & dynamic mechanisms of RNNs?
1/2 Our #NeurIPS2023 spotlight on a highly efficient algo for locating all fixed points, cycles, and bifurcation manifolds in RNNs: https://t.co/qmINjJB40h
@NeurIPSConf
By brilliant @Zahra__Monfared, @lukasironman, Nic Göring
Reconstructing computational system dynamics from neural data with recurrent neural networks — a Perspective by Daniel Durstewitz, Georgia Koppe & Max Ingo Thurm
https://t.co/0U3d5OTxxg
@DurstewitzLab
Reconstructing computational system dynamics from neural data with recurrent neural networks — a Perspective by Daniel Durstewitz, Georgia Koppe & Max Ingo Thurm
https://t.co/0U3d5OSZHI
@DurstewitzLab
Highlight of the week from @DurstewitzLab: Valuable insights on using Recurrent Neural Networks to study biological neural networks as dynamical systems!
https://t.co/td0xkuZ5dz
Excellent perspective/primer/review of 10 years of dynamical systems theory and apps to neuroscience, w/ glossary and 200+ citations.
"RNNs...may one day be transformative for our understanding of brain function, perhaps comparable in impact to the advances in optogenetics." 👀
In this article, @brenner_manuel dives into the fundamental mechanisms of several classes of generative models, shedding light on their inner workings and exploring their origins in and connections to neuroscience and cognition. https://t.co/X4t0zo8vNP
Our Perspective on reconstructing computat. system dynamics from neural data finally out in @NatRevNeurosci!
https://t.co/mje21ShyWC
We survey generative models that can be trained on time series to mimic the behavior of the neural substrate.
#AI#neuroscience#DynamicalSystems
We also discuss various ways of linking model latent spaces to the biological substrate through specific configuration of decoder models. This adds another layer of mechanistic interpretability to such models.
Jointly with @GeorgiaKoppe & @maxinthur
Find us tommorow, 9am at @CNSorg Leipzig in the Goethesaal for our tutorial on reconstructing dynamical systems from neural measurements using RNNs, including a hands-on tutorial with code!
Interested in investigating how neuronal dynamics supports the execution of cognitive functions?
Join us! 📢PhD position in the Brain Dynamics Laboratory at @SantAnnaPisa, Italy.
#manifold#RNN#compneuro
Deadline for the application: 4 August.
https://t.co/G7Y5aDrEtB
Interested in inferring interpretable, low-dim generative dynamical systems models from empirical time series? That after training also mimic the underlying system's long term behavior? Code from our #icml2023 oral now online:
https://t.co/GvQNuxlZD5
While many algos manage good short-term predictions, what is tricky is capturing the long-term dynamics. If you know of any other #ML/#AI/ #DataScience algos which can achieve this on real world data, please let us know!
Great work by Florian, @brenner_manuel, @Zahra__Monfared !