DurstewitzLab
@durstewitzlab.bsky.social
Scientific AI/ machine learning, dynamical systems (reconstruction), generative surrogate models of brains & behavior, applications in neuroscience & mental health
In our #NeurIPS2026 paper “A Minimal Interpretable Architecture for Zero-Shot Reconstruction of Dynamical Systems” (preprint: arxiv.org/abs/2607.14937) we reduce a DS FM to ingredients minimally necessary to reproduce long-term stat. and geom. properties of DS, even outperforming many TS & DS FM.
October 4, 2026 at 12:41 PM
In our #NeurIPS2026 paper “A Minimal Interpretable Architecture for Zero-Shot Reconstruction of Dynamical Systems” (preprint: arxiv.org/abs/2607.14937) we reduce a DS FM to ingredients minimally necessary to reproduce long-term stat. and geom. properties of DS, even outperforming many TS & DS FM.
In our #NeurIPS2026 paper “Topological Out-of-Domain Generalization in Dynamical Systems (DS) Reconstruction” (arxiv.org/abs/2606.22969) we aim to infer the DS generating observed TS jointly with control parameters, in order to extrapolate to different dynamical regimes, e.g. across tipping points.
October 2, 2026 at 2:40 PM
In our #NeurIPS2026 paper “Topological Out-of-Domain Generalization in Dynamical Systems (DS) Reconstruction” (arxiv.org/abs/2606.22969) we aim to infer the DS generating observed TS jointly with control parameters, in order to extrapolate to different dynamical regimes, e.g. across tipping points.
In our #NeurIPS2026 spotlight “Parallel-in-Time Training of Recurrent Neural Networks for Dynamical Systems (DS) Reconstruction” (preprint: arxiv.org/abs/2605.12683) we speed up training of nonlinear RNNs on time series from chaotic DS by >100x by combining DEER with generalized teacher forcing.
October 1, 2026 at 12:29 PM
In our #NeurIPS2026 spotlight “Parallel-in-Time Training of Recurrent Neural Networks for Dynamical Systems (DS) Reconstruction” (preprint: arxiv.org/abs/2605.12683) we speed up training of nonlinear RNNs on time series from chaotic DS by >100x by combining DEER with generalized teacher forcing.
Next week I will talk about *foundation models for dynamical systems reconstruction* at ECML in Naples, ml4its.github.io/ml4its2026/. I will cover some of the latest from the group, like efficient parallel-in-time training for DSR and minimal mechanisms for zero-shot DSR.
August 31, 2026 at 2:48 PM
Next week I will talk about *foundation models for dynamical systems reconstruction* at ECML in Naples, ml4its.github.io/ml4its2026/. I will cover some of the latest from the group, like efficient parallel-in-time training for DSR and minimal mechanisms for zero-shot DSR.
Reposted by DurstewitzLab
We congratulate @matthijspals.bsky.social on his successful PhD defense!
During his time in the Mackelab, he used RNNs to link neural activity with underlying mechanisms. Now he moved on to a Postdoc position in @durstewitzlab.bsky.social at the ZI Mannheim and the University of Heidelberg.
During his time in the Mackelab, he used RNNs to link neural activity with underlying mechanisms. Now he moved on to a Postdoc position in @durstewitzlab.bsky.social at the ZI Mannheim and the University of Heidelberg.
July 2, 2026 at 1:23 PM
We congratulate @matthijspals.bsky.social on his successful PhD defense!
During his time in the Mackelab, he used RNNs to link neural activity with underlying mechanisms. Now he moved on to a Postdoc position in @durstewitzlab.bsky.social at the ZI Mannheim and the University of Heidelberg.
During his time in the Mackelab, he used RNNs to link neural activity with underlying mechanisms. Now he moved on to a Postdoc position in @durstewitzlab.bsky.social at the ZI Mannheim and the University of Heidelberg.
In a #ICML2026 position paper we argue a dynamical systems perspective is needed to drive time series models forward: arxiv.org/abs/2602.16864
For TS, we need to move away from transformers that do not respect a system’s dynamical structure, esp. if out-of-domain generalization & insight is sought.
For TS, we need to move away from transformers that do not respect a system’s dynamical structure, esp. if out-of-domain generalization & insight is sought.
May 10, 2026 at 5:10 AM
In a #ICML2026 position paper we argue a dynamical systems perspective is needed to drive time series models forward: arxiv.org/abs/2602.16864
For TS, we need to move away from transformers that do not respect a system’s dynamical structure, esp. if out-of-domain generalization & insight is sought.
For TS, we need to move away from transformers that do not respect a system’s dynamical structure, esp. if out-of-domain generalization & insight is sought.
Neural ODEs are great as continuous-time dynamical systems, but slow and tedious to train.
In a new #ICML2026 paper we intro a novel solver for continuous-time RNNs that does not rely on numerical integration. It's not only way faster & robust, but enables explicit analysis: arxiv.org/abs/2602.15649
In a new #ICML2026 paper we intro a novel solver for continuous-time RNNs that does not rely on numerical integration. It's not only way faster & robust, but enables explicit analysis: arxiv.org/abs/2602.15649
May 10, 2026 at 4:58 AM
Neural ODEs are great as continuous-time dynamical systems, but slow and tedious to train.
In a new #ICML2026 paper we intro a novel solver for continuous-time RNNs that does not rely on numerical integration. It's not only way faster & robust, but enables explicit analysis: arxiv.org/abs/2602.15649
In a new #ICML2026 paper we intro a novel solver for continuous-time RNNs that does not rely on numerical integration. It's not only way faster & robust, but enables explicit analysis: arxiv.org/abs/2602.15649
Reposted by DurstewitzLab
i need the “llms are might be conscious” folx to read this
I met with AI data labelers in Kenya who are organizing their colleagues to fight the brutal working conditions and horrible pay given to the workers at the "bottom of the AI supply chain." They believe the NDAs they've signed are unenforceable so are speaking out:
www.404media.co/ai-is-africa...
www.404media.co/ai-is-africa...
'AI Is African Intelligence': The Workers Who Train AI Are Fighting Back
Kenyan workers are still the underpaid labor behind AI training, moderation, and sex chatbots. The Data Labelers Association is fighting back.
www.404media.co
March 14, 2026 at 1:06 PM
i need the “llms are might be conscious” folx to read this
Reposted by DurstewitzLab
From the top of my head, here some recent ones:
"Two views on the cognitive brain" by @johnwkrakauer.bsky.social, @dlbarack.bsky.social
"Reconstructing computational system dynamics from neural data with recurrent neural networks" by @durstewitzlab.bsky.social et al
1/3
"Two views on the cognitive brain" by @johnwkrakauer.bsky.social, @dlbarack.bsky.social
"Reconstructing computational system dynamics from neural data with recurrent neural networks" by @durstewitzlab.bsky.social et al
1/3
I’m building a foundational reading list for our lab (systems & circuit neuroscience, compneuro, modeling, neuromodulators, population coding etc.).
I’d like to crowdsource recommendations.
Which review(s) would you consider mandatory reading for the next generation of researchers?
I’d like to crowdsource recommendations.
Which review(s) would you consider mandatory reading for the next generation of researchers?
March 1, 2026 at 2:28 PM
From the top of my head, here some recent ones:
"Two views on the cognitive brain" by @johnwkrakauer.bsky.social, @dlbarack.bsky.social
"Reconstructing computational system dynamics from neural data with recurrent neural networks" by @durstewitzlab.bsky.social et al
1/3
"Two views on the cognitive brain" by @johnwkrakauer.bsky.social, @dlbarack.bsky.social
"Reconstructing computational system dynamics from neural data with recurrent neural networks" by @durstewitzlab.bsky.social et al
1/3
In a new #ICLR2026 paper we provide an algorithm for semi-analytically constructing un-/stable manifolds of fixed points and cycles of ReLU-based RNNs:
openreview.net/pdf?id=EAwLA...
These manifolds provide a skeleton for the system’s dynamics, dissecting the state space into basins of attraction.
openreview.net/pdf?id=EAwLA...
These manifolds provide a skeleton for the system’s dynamics, dissecting the state space into basins of attraction.
March 1, 2026 at 4:57 PM
In a new #ICLR2026 paper we provide an algorithm for semi-analytically constructing un-/stable manifolds of fixed points and cycles of ReLU-based RNNs:
openreview.net/pdf?id=EAwLA...
These manifolds provide a skeleton for the system’s dynamics, dissecting the state space into basins of attraction.
openreview.net/pdf?id=EAwLA...
These manifolds provide a skeleton for the system’s dynamics, dissecting the state space into basins of attraction.
We had a go at a blog about our recent dynamical systems foundation model published at NeurIPS (with strong support from the Structures outreach team!) … let us know your thoughts!
With DynaMix, researchers bring ideas from physics into machine learning to uncover the dynamics underlying complex data.
Learn all about this on the new post by Daniel Durstewitz @durstewitzlab.bsky.social and Christoph Hemmer: structures.uni-heidelberg.de/blog/posts/2...
2/2
#Dynamics #AI
Learn all about this on the new post by Daniel Durstewitz @durstewitzlab.bsky.social and Christoph Hemmer: structures.uni-heidelberg.de/blog/posts/2...
2/2
#Dynamics #AI
STRUCTURES Blog | Predicting the Future: From Cave Paintings to DynaMix
Blog of the STRUCTURES Cluster of Excellence. On this blog we show pictures of interesting structures related to our research in the STRUCTURES Cluster, and we give you an idea what they are, how they...
structures.uni-heidelberg.de
February 19, 2026 at 6:26 PM
We had a go at a blog about our recent dynamical systems foundation model published at NeurIPS (with strong support from the Structures outreach team!) … let us know your thoughts!
Reposted by DurstewitzLab
Fully-funded International Neuroscience Doctoral Programme🧠 Champalimaud Foundation, Lisbon, Portugal 🇵🇹
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
December 16, 2025 at 7:20 PM
Fully-funded International Neuroscience Doctoral Programme🧠 Champalimaud Foundation, Lisbon, Portugal 🇵🇹
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
Deadline: Jan 31, 2026
fchampalimaud.org/champalimaud...
Research program spans systems/computational/theoretical/clinical/sensory/motor neuroscience, neuroethology, intelligence, and more!!
Tomorrow Christoph will present DynaMix, the first foundation model for dynamical systems reconstruction, at #NeurIPS2025 Exhibit Hall C,D,E #2303
December 5, 2025 at 1:28 PM
Tomorrow Christoph will present DynaMix, the first foundation model for dynamical systems reconstruction, at #NeurIPS2025 Exhibit Hall C,D,E #2303
Unlike current AI systems, animals can quickly and flexibly adapt to changing environments.
This is the topic of our new perspective in Nature MI (rdcu.be/eSeif), where we relate dynamical and plasticity mechanisms in the brain to in-context and continual learning in AI. #NeuroAI
This is the topic of our new perspective in Nature MI (rdcu.be/eSeif), where we relate dynamical and plasticity mechanisms in the brain to in-context and continual learning in AI. #NeuroAI
What neuroscience can tell AI about learning in continuously changing environments
Nature Machine Intelligence - Durstewitz et al. explore what artificial intelligence can learn from the brain’s ability to adjust quickly to changing environments. By linking neuroscience...
rdcu.be
November 29, 2025 at 9:24 AM
Unlike current AI systems, animals can quickly and flexibly adapt to changing environments.
This is the topic of our new perspective in Nature MI (rdcu.be/eSeif), where we relate dynamical and plasticity mechanisms in the brain to in-context and continual learning in AI. #NeuroAI
This is the topic of our new perspective in Nature MI (rdcu.be/eSeif), where we relate dynamical and plasticity mechanisms in the brain to in-context and continual learning in AI. #NeuroAI
Revised version of our #NeurIPS2025 paper with full code base in Julia & Python now online, see arxiv.org/abs/2505.13192
October 28, 2025 at 6:27 PM
Revised version of our #NeurIPS2025 paper with full code base in Julia & Python now online, see arxiv.org/abs/2505.13192
Our #AI #DynamicalSystems #FoundationModel DynaMix was accepted to #NeurIPS2025 with outstanding reviews (6555) – first model which can *zero-shot*, w/o any fine-tuning, forecast the *long-term statistics* of time series provided a context. Test it on #HuggingFace:
huggingface.co/spaces/Durst...
huggingface.co/spaces/Durst...
September 21, 2025 at 9:40 AM
Our #AI #DynamicalSystems #FoundationModel DynaMix was accepted to #NeurIPS2025 with outstanding reviews (6555) – first model which can *zero-shot*, w/o any fine-tuning, forecast the *long-term statistics* of time series provided a context. Test it on #HuggingFace:
huggingface.co/spaces/Durst...
huggingface.co/spaces/Durst...
We have openings for several fully-funded positions (PhD & PostDoc) at the intersection of AI/ML, dynamical systems, and neuroscience within a BMFTR-funded Neuro-AI consortium, at Heidelberg University & Central Institute of Mental Health:
www.einzigartigwir.de/en/job-offer...
More info below ...
www.einzigartigwir.de/en/job-offer...
More info below ...
August 15, 2025 at 7:46 AM
We have openings for several fully-funded positions (PhD & PostDoc) at the intersection of AI/ML, dynamical systems, and neuroscience within a BMFTR-funded Neuro-AI consortium, at Heidelberg University & Central Institute of Mental Health:
www.einzigartigwir.de/en/job-offer...
More info below ...
www.einzigartigwir.de/en/job-offer...
More info below ...
Reposted by DurstewitzLab
Is it possible to go from spikes to rates without averaging?
We show how to exactly map recurrent spiking networks into recurrent rate networks, with the same number of neurons. No temporal or spatial averaging needed!
Presented at Gatsby Neural Dynamics Workshop, London.
We show how to exactly map recurrent spiking networks into recurrent rate networks, with the same number of neurons. No temporal or spatial averaging needed!
Presented at Gatsby Neural Dynamics Workshop, London.
From Spikes To Rates
YouTube video by Gerstner Lab
youtu.be
August 8, 2025 at 3:25 PM
Is it possible to go from spikes to rates without averaging?
We show how to exactly map recurrent spiking networks into recurrent rate networks, with the same number of neurons. No temporal or spatial averaging needed!
Presented at Gatsby Neural Dynamics Workshop, London.
We show how to exactly map recurrent spiking networks into recurrent rate networks, with the same number of neurons. No temporal or spatial averaging needed!
Presented at Gatsby Neural Dynamics Workshop, London.
Reposted by DurstewitzLab
Today I joined >1900 members of US National Academies of Science, Engineering & Medicine signing this open letter (views our own).
Leadership of science by US has been paramount for >70yrs & Admin is now acting to throw it all away!
docs.google.com/document/d/1...
www.nytimes.com/2025/03/31/s...
Leadership of science by US has been paramount for >70yrs & Admin is now acting to throw it all away!
docs.google.com/document/d/1...
www.nytimes.com/2025/03/31/s...
Public Statement on Supporting Science for the Benefit of All Citizens
TO THE AMERICAN PEOPLE We all rely on science. Science gave us the smartphones in our pockets, the navigation systems in our cars, and life-saving medical care. We count on engineers when we drive acr...
docs.google.com
March 31, 2025 at 5:00 PM
Today I joined >1900 members of US National Academies of Science, Engineering & Medicine signing this open letter (views our own).
Leadership of science by US has been paramount for >70yrs & Admin is now acting to throw it all away!
docs.google.com/document/d/1...
www.nytimes.com/2025/03/31/s...
Leadership of science by US has been paramount for >70yrs & Admin is now acting to throw it all away!
docs.google.com/document/d/1...
www.nytimes.com/2025/03/31/s...
Reposted by DurstewitzLab
What a fantastic accomplishment -- and what a fantastic story! www.quantamagazine.org/at-17-hannah...
At 17, Hannah Cairo Solved a Major Math Mystery | Quanta Magazine
After finding the homeschooling life confining, the teen petitioned her way into a graduate class at Berkeley, where she ended up disproving a 40-year-old conjecture.
www.quantamagazine.org
August 3, 2025 at 12:13 PM
What a fantastic accomplishment -- and what a fantastic story! www.quantamagazine.org/at-17-hannah...
Got prov. approval for 2 major grants in Neuro-AI & Dynamical Systems Reconstruction, on learning & inference in non-stationary environments, out-of-domain generalization, and DS foundation models. To all AI/math/DS enthusiasts: Expect job announcements (PhD/PostDoc) soon! Feel free to get in touch.
July 13, 2025 at 6:23 AM
Got prov. approval for 2 major grants in Neuro-AI & Dynamical Systems Reconstruction, on learning & inference in non-stationary environments, out-of-domain generalization, and DS foundation models. To all AI/math/DS enthusiasts: Expect job announcements (PhD/PostDoc) soon! Feel free to get in touch.
We wrote a little #NeuroAI piece about in-context learning & neural dynamics vs. continual learning & plasticity, both mechanisms to flexibly adapt to changing environments:
arxiv.org/abs/2507.02103
We relate this to non-stationary rule learning tasks with rapid performance jumps.
Feedback welcome!
arxiv.org/abs/2507.02103
We relate this to non-stationary rule learning tasks with rapid performance jumps.
Feedback welcome!
What Neuroscience Can Teach AI About Learning in Continuously Changing Environments
Modern AI models, such as large language models, are usually trained once on a huge corpus of data, potentially fine-tuned for a specific task, and then deployed with fixed parameters. Their training ...
arxiv.org
July 6, 2025 at 10:18 AM
We wrote a little #NeuroAI piece about in-context learning & neural dynamics vs. continual learning & plasticity, both mechanisms to flexibly adapt to changing environments:
arxiv.org/abs/2507.02103
We relate this to non-stationary rule learning tasks with rapid performance jumps.
Feedback welcome!
arxiv.org/abs/2507.02103
We relate this to non-stationary rule learning tasks with rapid performance jumps.
Feedback welcome!
Happy to discuss our work on parsimonious & math. tractable RNNs for dynamical systems reconstruction next week at
cns2025florence.sched.com/event/1z9Mt/...
cns2025florence.sched.com/event/1z9Mt/...
CNS*2025 Florence: NeuroXAI: Explainable AI for Understandi...
View more about this event at CNS*2025 Florence
cns2025florence.sched.com
July 3, 2025 at 12:40 PM
Happy to discuss our work on parsimonious & math. tractable RNNs for dynamical systems reconstruction next week at
cns2025florence.sched.com/event/1z9Mt/...
cns2025florence.sched.com/event/1z9Mt/...
How do animals learn new rules? By systematically testing diff. behavioral strategies, guided by selective attn. to rule-relevant cues: rdcu.be/etlRV
Akin to in-context learning in AI, strategy selection depends on the animals' "training set" (prior experience), with similar repr. in rats & humans.
Akin to in-context learning in AI, strategy selection depends on the animals' "training set" (prior experience), with similar repr. in rats & humans.
Abstract rule learning promotes cognitive flexibility in complex environments across species
Nature Communications - Whether neurocomputational mechanisms that speed up human learning in changing environments also exist in other species remains unclear. Here, the authors show that both...
rdcu.be
June 26, 2025 at 3:30 PM
How do animals learn new rules? By systematically testing diff. behavioral strategies, guided by selective attn. to rule-relevant cues: rdcu.be/etlRV
Akin to in-context learning in AI, strategy selection depends on the animals' "training set" (prior experience), with similar repr. in rats & humans.
Akin to in-context learning in AI, strategy selection depends on the animals' "training set" (prior experience), with similar repr. in rats & humans.
Reposted by DurstewitzLab
What a line up!! With Lorenzo Gaetano Amato, Demian Battaglia, @durstewitzlab.bsky.social, @engeltatiana.bsky.social, @seanfw.bsky.social, Matthieu Gilson, Maurizio Mattia, @leonardopollina.bsky.social, Sara Solla.
June 21, 2025 at 10:24 AM
What a line up!! With Lorenzo Gaetano Amato, Demian Battaglia, @durstewitzlab.bsky.social, @engeltatiana.bsky.social, @seanfw.bsky.social, Matthieu Gilson, Maurizio Mattia, @leonardopollina.bsky.social, Sara Solla.