Taku Ito
@takuito.bsky.social
📤 577
📥 534
📝 5
Computational Neuroscience + AI @ IBM Research | 📍NYC |
https://ito-takuya.github.io
pinned post!
What complexity of algorithms can AI compute? In a new paper with colleagues at IBM Research, we explore how circuit complexity theory can help quantify the degree of algorithmic generalization in AI systems.
www.nature.com/articles/s42...
@natmachintell.nature.com
#ML
#AI
#MLSky
1/n
loading . . .
3 months ago
1
16
6
reposted by
Taku Ito
Mary Elizabeth Sutherland
24 days ago
Did you know that AI can figure out its own way to learn, and that its way is better than one designed by humans? Read more in a
@nature.com
N&V (and the original paper is in the comment) 🧪
www.nature.com/articles/d41...
loading . . .
AI discovers learning algorithm that outperforms those designed by humans
An artificial-intelligence algorithm that discovers its own way to learn achieves state-of-the-art performance, including on some tasks it had never encountered before.
https://www.nature.com/articles/d41586-025-03398-6
2
6
2
reposted by
Taku Ito
Tatiana Engel
23 days ago
Our work with
@pawa-pawa.bsky.social
is out in Nature Machine Intelligence! The choice of activation function affects the representations, dynamics, and circuit solutions that emerge in RNNs trained on cognitive tasks. Activation matters!
www.nature.com/articles/s42...
add a skeleton here at some point
0
39
9
reposted by
Taku Ito
Ramon Astudillo
about 1 month ago
(repost welcome) The Generative Model Alignment team at IBM Research is looking for next summer interns! Two candidates for two topics 🍰Reinforcement Learning environments for LLMs 🐎Speculative and non-auto regressive generation for LLMs interested/curious? DM or email
[email protected]
1
19
15
reposted by
Taku Ito
PessoaBrain
about 1 month ago
Michael X Cohen on why he left academia/neuroscience.
mikexcohen.substack.com/p/why-i-left...
loading . . .
Why I left academia and neuroscience
Don't worry, this isn't yet another story of rage-quitting.
https://mikexcohen.substack.com/p/why-i-left-academia-and-neuroscience
7
94
50
reposted by
Taku Ito
Nature
about 2 months ago
Nature research paper: Arousal as a universal embedding for spatiotemporal brain dynamics
go.nature.com/4nMUgYz
loading . . .
Arousal as a universal embedding for spatiotemporal brain dynamics - Nature
Reframing of arousal as a latent dynamical system can reconstruct multidimensional measurements of large-scale spatiotemporal brain dynamics on the timescale of seconds in mice.
https://go.nature.com/4nMUgYz
0
29
13
reposted by
Taku Ito
Michael W. Cole
2 months ago
Lab’s latest is out in Imaging Neuroscience, led by Kirsten Peterson: “Regularized partial correlation provides reliable functional connectivity estimates while correcting for widespread confounding”, where we demonstrate a major improvement to standard fMRI functional connectivity (correlation) 1/n
5
74
30
What complexity of algorithms can AI compute? In a new paper with colleagues at IBM Research, we explore how circuit complexity theory can help quantify the degree of algorithmic generalization in AI systems.
www.nature.com/articles/s42...
@natmachintell.nature.com
#ML
#AI
#MLSky
1/n
loading . . .
3 months ago
1
16
6
reposted by
Taku Ito
8 months ago
Mental health research is at a turning point—breakthroughs can transform lives, but only with bold action, investment, and open collaboration. The time for action is now. Read our full statement here:
childmind.org/blog/can-sci...
0
15
7
reposted by
Taku Ito
Guy Davidson
9 months ago
Out today in Nature Machine Intelligence! From childhood on, people can create novel, playful, and creative goals. Models have yet to capture this ability. We propose a new way to represent goals and report a model that can generate human-like goals in a playful setting... 1/N
5
135
44
reposted by
Taku Ito
Naoki Hiratani
9 months ago
New preprint! Ziyan and I explore how task order impacts continual learning in neural networks and how to optimize it. Our analysis highlights two key principles for better task sequencing. Check it out:
arxiv.org/pdf/2502.03350
loading . . .
https://arxiv.org/pdf/2502.03350
0
7
3
reposted by
Taku Ito
Dr. Becca
10 months ago
The entire website for the NIH Office of Research on Women's Health (ORWH) is very nearly stripped bare. This is so, so devastating.
orwh.od.nih.gov/research/fun...
loading . . .
https://orwh.od.nih.gov/research/funded-research-and-programs/funding-opportunities-and-notices
44
982
686
reposted by
Taku Ito
Julien Corbo
10 months ago
New paper out! 🚨 📰 With
@batuhanerkat.bsky.social
, John McClure,
@hussainyk1.bsky.social
,
@polacklab.bsky.social
we reveal how discretized representations in V1 predict suboptimal orientation discrimination. 🧪🧠🐭 This work reconciles neuro and psychometric curves
www.nature.com/articles/s41...
loading . . .
Discretized representations in V1 predict suboptimal orientation discrimination - Nature Communications
How animals generate perceptual decisions remains poorly understood. Here, the authors show that during a discrimination task, the mouse visual cortex does not encode the orientations of the cues but ...
https://www.nature.com/articles/s41467-024-55409-1
3
26
8
reposted by
Taku Ito
Veith Weilnhammer, MD
10 months ago
New paper in
@brain1878.bsky.social
: Healthy people under S-ketamine, an NMDAR antagonist, and people living with schizophrenia, a disorder associated with NMDAR hypofunction, spend more time in an external mode of perception - where noisy sensory signals override knowledge about the world.
1
26
10
reposted by
Taku Ito
Joey Saito
10 months ago
The origin of color categories | PNAS
www.pnas.org/doi/10.1073/...
loading . . .
The origin of color categories | PNAS
To what extent does concept formation require language? Here, we exploit color to address this question and ask whether macaque monkeys have color ...
https://www.pnas.org/doi/10.1073/pnas.2400273121
2
53
19
reposted by
Taku Ito
Joao Barbosa
10 months ago
Check our latest in which we leverage shape metrics to compare neural geometry across regions, sessions or subjects and how their differences predict behavior. w/ Nejatbakhsh, Duong,
@sarah-harvey.bsky.social
, Brincat,
@siegellab.bsky.social
,
@earlkmiller.bsky.social
&
@itsneuronal.bsky.social
add a skeleton here at some point
3
103
38
reposted by
Taku Ito
Ethan Mollick
10 months ago
Paper shows very small LLMs can match or beat larger ones through 'deep thinking' - evaluating different solution paths - and other tricks. Their 7B model beats o1-preview on complex math by exploring 64 different solutions & picking the best one. Test-time compute paradigm seems really fruitful.
3
157
24
reposted by
Taku Ito
Marlene Cohen
11 months ago
New results for a new year! “Linking neural population formatting to function” describes our modern take on an old question: how can we understand the contribution of a brain area to behavior?
www.biorxiv.org/content/10.1...
🧠👩🏻🔬🧪🧵
#neuroskyence
1/
loading . . .
Linking neural population formatting to function
Animals capable of complex behaviors tend to have more distinct brain areas than simpler organisms, and artificial networks that perform many tasks tend to self-organize into modules (1-3). This sugge...
https://www.biorxiv.org/content/10.1101/2025.01.03.631242v1
2
231
88
reposted by
Taku Ito
Blake Richards
11 months ago
And relatedly, Felix wrote a good piece on the stress and anxiety currently affecting many people who work in AI due to the current climate in the industry:
docs.google.com/document/d/1...
If only more folks in AI were gentle and introspective like this...
loading . . .
AI and Stress
200Bn Weights of Responsibility The Stress of Working in Modern AI Felix Hill, Oct 2024 The field of AI has changed irrevocably in the last 2 years. ChatGPT is approaching 200m monthly users. Gemin...
https://docs.google.com/document/d/1aEdTE-B6CSPPeUWYD-IgNVQVZM25f7MF-u9qn5KJJvo/mobilebasic
0
17
5
reposted by
Taku Ito
David Bau
11 months ago
What was the most important machine learning paper in 2024? My Famous Deep Learning Papers list (that I use in teaching) does not include any new ideas from the last year.
papers.baulab.info
Which single new paper would you add?
10
55
11
reposted by
Taku Ito
Melanie Mitchell
11 months ago
Some of my thoughts on OpenAI's o3 and the ARC-AGI benchmark
aiguide.substack.com/p/did-openai...
loading . . .
Did OpenAI Just Solve Abstract Reasoning?
OpenAI’s o3 model aces the "Abstraction and Reasoning Corpus" — but what does it mean?
https://aiguide.substack.com/p/did-openai-just-solve-abstract-reasoning
17
341
125
reposted by
Taku Ito
Xiaoxuan
11 months ago
📌 Poster Session: ⏰ When: TODAY, Thu, Dec 12, 4:30 p.m. – 7:30 p.m. PST 📍 Where: East Exhibit Hall A-C, #3705 📄 What: Geometry of Naturalistic Object Representations in Recurrent Neural Network Models of Working Memory Hope to see you there!
@bashivan.bsky.social
@takuito.bsky.social
0
3
2
reposted by
Taku Ito
Jean-Rémi King
11 months ago
🚨We're very excited to share our latest study, by Pablo Diego and team: "A polar coordinate system represents syntax in large language models", 📄: Paper
arxiv.org/abs/2412.05571
🪧: Poster tomorrow:
neurips.cc/virtual/2024..
. 🧵: Thread 👇
loading . . .
2
58
18
reposted by
Taku Ito
Adeel Razi
11 months ago
Just published🔈 "Structurally informed models of directed brain connectivity" Read:
rdcu.be/d3dC4
We review how structural connectivity constrains directed connectivity models 🧠 Lead by
@matthewdgreaves.bsky.social
w/
@novelli-leo.bsky.social
,
@sinamansourl.bsky.social
and Andrew Zalesky
1
37
15
reposted by
Taku Ito
Rick Betzel
12 months ago
hey -- i'm hiring a postdoc! the ad will be up shortly, but looking for someone with network neuroscience experience (very broadly). the position isn't tied to any specific project/grant, so lots of flexibility in terms of what you'd actually *do*. hmu if you might be interested/want to learn more!
1
52
51
reposted by
Taku Ito
Michael W. Cole
12 months ago
“Cognitive flexibility as the shifting of brain network flows by flexible neural representations”, a solo paper by yours truly, making the case that brain activity flow shifts are essential to mental flexibility (and quite interesting too!) Open access:
www.sciencedirect.com/science/arti...
2
102
26
Our new NeurIPS paper on naturalistic representations in dynamic WM models, led by
@xiaoxuanlei.bsky.social
and
@bashivan.bsky.social
Thread by Xiaoxuan 👇
add a skeleton here at some point
12 months ago
1
7
3
reposted by
Taku Ito
CLaE
12 months ago
JAMA Psychiatry A Dynamical Systems View of Psychiatric Disorders—Theory A Review
jamanetwork.com/journals/jam...
loading . . .
A Dynamical Systems View of Psychiatric Disorders—Theory
This narrative review describes a new approach to the diagnosis and treatment of psychiatric disorders that is based on dynamical systems theory, which addresses the concepts of tipping points, cycles...
https://jamanetwork.com/journals/jamapsychiatry/article-abstract/2817087
3
28
10
reposted by
Taku Ito
Michael W. Cole
about 1 year ago
Lab’s latest at PLOS Comp Biol, led by Carrisa Cocuzza: “Distributed network flows generate localized category selectivity in human visual cortex”. This one changed how I think the brain works! Even "localized" functions are likely generated by distributed processes
doi.org/10.1371/jour...
loading . . .
Distributed network flows generate localized category selectivity in human visual cortex
Author summary A fundamental question in neuroscience has persisted for over a century: to what extent do distributed processes drive brain function? The existence of category-selective regions within...
https://doi.org/10.1371/journal.pcbi.1012507
0
17
12
reposted by
Taku Ito
bioRxiv Neuroscience
about 1 year ago
Practice Reshapes the Geometry and Dynamics of Task-tailored Representations https://www.biorxiv.org/content/10.1101/2024.09.12.612718v1
loading . . .
Practice Reshapes the Geometry and Dynamics of Task-tailored Representations https://www.biorxiv.org/content/10.1101/2024.09.12.612718v1
Extensive practice makes task performance more efficient and precise, leading to automaticity. Howev
https://www.biorxiv.org/content/10.1101/2024.09.12.612718v1
0
1
1
reposted by
Taku Ito
Kohitij Kar
almost 2 years ago
This new article perfectly concludes my time & lessons in Jim’s lab.@JamesJDiCarlo and I propose +review SMART models of object recognition ✅ Sensory computable ✅ Mechanistic ✅ Anatomically Referenced ✅ Testable Coming in Annual Reviews 2024 Preprint:
bit.ly/3tk7u8D
0
8
4
reposted by
Taku Ito
hakwan lau
about 2 years ago
www.nature.com/articles/s41...
Lake & Baroni "Fodor and Pylyshyn famously argued that artificial neural networks ... are ... not viable models of the mind.... Here we [show] that neural networks can achieve human-like systematicity when optimized for their compositional skills."
0
16
10
reposted by
Taku Ito
Earl K. Miller
about 2 years ago
From lazy to rich to exclusive task representations in neural networks and neural codes
doi.org/10.1016/j.co...
#neuroscience
0
16
11
reposted by
Taku Ito
CLaE
about 2 years ago
NeuroImage Spatially heterogeneous structure-function coupling in haemodynamic and electromagnetic brain networks
www.sciencedirect.com/science/arti...
loading . . .
Spatially heterogeneous structure-function coupling in haemodynamic and electromagnetic brain networ...
The relationship between structural and functional connectivity in the brain is a key question in connectomics. Here we quantify patterns of structure…
https://www.sciencedirect.com/science/article/pii/S1053811923004275?via%3Dihub
0
1
1
reposted by
Taku Ito
Jeff Johnston
about 2 years ago
How does the brain represent multiple different things at once in a single population of neurons?
@justfineneuro.bsky.social
,
@benhayden.bsky.social
, B Ebitz, M Yoo, and I show that it uses semi-orthogonal subspaces for each item. Preprint here:
arxiv.org/abs/2309.07766
Clouds below! (1/n)
loading . . .
Semi-orthogonal subspaces for value mediate a tradeoff between...
When choosing between options, we must associate their values with the action needed to select them. We hypothesize that the brain solves this binding problem through neural population subspaces....
https://arxiv.org/abs/2309.07766
1
28
12
reposted by
Taku Ito
CLaE
about 2 years ago
Semi-orthogonal subspaces for value mediate a tradeoff between binding and generalization
arxiv.org/abs/2309.07766
loading . . .
Semi-orthogonal subspaces for value mediate a tradeoff between...
When choosing between options, we must associate their values with the action needed to select them. We hypothesize that the brain solves this binding problem through neural population subspaces....
https://arxiv.org/abs/2309.07766
0
9
2
you reached the end!!
feeds!
log in