@NeuroGoudar
Joined October 2017
  • Tweets17
  • Following185
  • Followers66
Vishwa Goudar retweeted
Wayne Soo@soowmwayne11 Oct 2023
It is challenging to train continuous-time biological RNN models to learn long-term dependencies. In our latest paper (in #NeurIPS2023) with @NeuroGoudar @xjwanglab, we propose several solutions. 1/8 biorxiv.org/content/10.1101/…

Training biologically plausible recurrent neural networks on cognitive tasks with long-term...

Training recurrent neural networks (RNNs) has become a go-to approach for generating and evaluating mechanistic neural hypotheses for cognition. The ease and efficiency of training RNNs with backpr...

biorxiv.org
1
18
58
4,913
Our new paper is about the most scientifically bewildering state in meditation. If we figure this one out it’s a goldmine for consciousness research, possibly revealing about human hibernation, or useful for space travel. Who knows. Here's the wild story 👇
https://shorturl.at/dkPRY

ALT https://shorturl.at/dkPRY

57
298
89
1,590
273,616
Schema formation in a neural population subspace underlies learning-to-learn in flexible sensorimotor problem-solving nature.com/articles/s41593-0…

Schema formation in a neural population subspace underlies learning-to-learn in flexible sensorim...

Nature Neuroscience - Learning-to-learn refers to progressive speedup in solving a series of problems with shared structure. This study shows that it emerges in recurrent neural networks from the...

nature.com
13
40
14,472
Vishwa Goudar retweeted
bioRxiv@biorxivpreprint11 Jan 2023
Comparing rapid rule-learning strategies in humans and monkeys biorxiv.org/cgi/content/shor… #bioRxiv
1
2
3
2,537
Vishwa Goudar retweeted
Dean Buono@DeanBuono11 Jul 2022
Replying to @shanglin_zhou
@shanglin_zhou Mike Seay @JiannisTax @pgolshani introduce a differential-Delay-Match-Sample task, and show that human performance is influenced by the task irrelevant delays. RNNs that capture human performance multiplex WM and time in neural sequences. biorxiv.org/content/10.1101/…

Multiplexing working memory and time: encoding retrospective and prospective information in neural...

Working memory (WM) and timing are generally considered distinct cognitive functions, but similar neural signatures have been implicated in both. To explore the hypothesis that WM and timing may rely...

biorxiv.org
7
1
14
With @rodlaje and @HMotanis we prove that classic homeostatic rules -when applied to all synapse types- are unstable and cannot lead to inhibition-stabilized regimes such as Up-states. We report a family of rules that autonomously lead to stable Up-states and balanced dynamics.
bioRxiv Neuroscience@biorxiv_neursci3 Jan 2021
Orchestrated Excitatory and Inhibitory Learning Rules Lead to the Unsupervised Emergence of Up-states and Balanced Network Dynamics biorxiv.org/cgi/content/shor… #biorxiv_neursci
1
3
1
7
Thrilled to share that our new paper discussing network dynamics in #FragileX is online nature.com/articles/s41598-0… We show the following:
6
6
1
31
Vishwa Goudar retweeted
Dave Freedman@djfreedman26 Mar 2022
Gordon Conference on the Neurobiology of Cognition is on for this summer! Please join and spread the word! July 24-29, 2022 at Sunday Rivery in Newry, Maine. Apply now to attend and present your work! Trainees, please attend the GRS immediately prior! grc.org/neurobiology-of-cogn…
1
26
62
Preprint w @johndmurray and @adredish, “Human latent-state generalization through prototype learning with discriminative attention.” This paper could interest those in neuroscience, psychology, psychiatry, or machine learning. psyarxiv.com/ku4fr
3
18
1
55
Out sciencedirect.com/science/ar… in Neuron today. This was a great collaboration in which Stefano Recanatesi and I are co-first authors working together with Luca Mazzucato and Zach Mainen labs. We combine electrophysiology, behavior, data analysis, and network modeling. 1/4
1
16
2
67
Swartz Seminar @NYU_CNS Tomorrow at 12pm: Prof Monty Escabi – "Encoding and perceiving the texture of sounds: auditory midbrain codes for recognizing and categorizing auditory texture and for listening in noise" | Fri, Oct 1, 2021 via @worldwideneuro world-wide.org/seminar/7602
4
4
Friday 17th September (tomorrow), Dr. Xiaosi Gu will give a talk "The Social Brain: From Model to Mental Health" for the Swartz Seminar (organized by myself and @AllenLiuYue) at @NYU_CNS The talk (and questions) can be followed online at: youtube.com/watch?v=f-bVDBFR…
6
6
Vishwa Goudar retweeted
Sara A Solla@SaraASolla4 Sep 2021
Low dimensional representations of population neural activity - dare I say manifolds - play a fundamental role in learning facilitation.
Δρ. Θεοδώνη@ptheodoni3 Sep 2021
"The neural substrate of a schema emerges within a low-dimensional subspace of population activity. Its reuse in new problems facilitates learning by limiting connection weight changes. ..learning is determined by the vector field change" Congrats @NeuroGoudar! Beautiful work
6
19
Elucidating the neural mechanisms of Learning-to-Learn biorxiv.org/cgi/content/shor… #biorxiv_neursci
12
1
41
Vishwa Goudar retweeted
Tim Vogels@TPVogels3 Sep 2020
For those of you who love seeing my face, I'll be on NYU's @worldwideneuro tomorrow, 18:00 Vienna, 12:00 NYC, worldwideneuro.com/seminar-e…. If you only want the science, join at 18:05 & stick a post-it onto the lower right screen. Oh, and here's a hint what I'll be talking about ;)
1
8
27
Vishwa Goudar retweeted
Wei Ji Ma@weijima0112 Oct 2018
I don't understand the technical side but the human story is fascinating: quantamagazine.org/graduate-… via @QuantaMagazine (1/3)
1
2
12
Load more