Meet Barot @mythosscience

Self-organizing neural nets and artificial life. PhD from @NYUdatascience meetbarot.com New York Joined December 2023

Tweets

70
Followers

83
Following

245
Likes

949

Meet Barot @mythosscience

5 months ago

Within the next 5 years there's going to be a better, more efficient way to train neural networks than gradient-based optimization. My bet is that it's going to take inspiration from living organisms, with properties like self-organization, replication, and evolution.

NYU Center for Data Science @NYUDataScience

5 months ago

0 0 1 528 0

Download Video

0 0 2 90 0

Nautilus Magazine @NautilusMag

5 months ago

Researchers are rediscovering the forgotten legacy of a pioneering Black scientist named Charles Henry Turner, who conducted trailblazing research on the cognitive traits of bees, spiders, and more. nautil.us/charles-henry-…

0 6 5 1K 2

CalCo @calco_io

6 months ago

Introducing a new state-of-the-art code embedding model - SOTA performance on the CodeSearchNet benchmark - Truly open (weights, data, code) - Apache 2.0 Nomic Embed Code brings us one step closer to a world with truly open source embedding models for every modality! 🧵

7 48 340 54K 175

Download Image

Meet Barot @mythosscience

9 months ago

Test time compute should be on weight generation instead of token generation

3 0 0 237 0

Meet Barot @mythosscience

10 months ago

Maybe different shapes of this landscape can be suited to different modes of life. the shape suited to society now may not always have been so suited to it before

Mitchell B. Slapik @mslapik

10 months ago

Maybe different shapes of this landscape can be suited to different modes of life. the shape suited to society now may not always have been so suited to it before

344 2K 15K 1.9M 8K

Download Gif

1 0 1 314 0

Meet Barot @mythosscience

11 months ago

i think the only way you don't develop agi exclusively for state power is if you have both a decentralized development and compute process. any other way will be captured by the state, there'd be too much at stake.

Daniel Faggella @danfaggella

11 months ago

8 10 80 4K 7

0 0 1 130 0

Bojan Tunguz @tunguz

11 months ago

This is honestly the biggest threat to the current AI - someone discovering how you can model all those trillions of coefficients with a relatively simple formula.

Hamid Naderi Yeganeh @naderi_yeganeh

11 months ago

This is honestly the biggest threat to the current AI - someone discovering how you can model all those trillions of coefficients with a relatively simple formula.

982 15K 163K 3.8M 16K

Download Image

40 40 761 88K 202

Meet Barot @mythosscience

12 months ago

kinda sense that we're gonna get more hopfield network papers now

0 0 2 70 0

Kenneth Stanley @kenneth0stanley

a year ago

Prediction: while backprop will likely play a role in any future AGI, there will also be at least one completely novel weight-updating learning algorithm that is yet to be invented today.

24 24 265 41K 78

afika @afikanyati

a year ago

@RobertHaisfield your mind already kind of does this. @drmichaellevin calls this the multi-scale competency architecture of the body. our cells have been shown to have a minimal level of cognition, so in a dynamic programming-like way, our bodies break down large problems into smaller ones that…