Lightnews — Scholar-powered news

Amine El Ouassouli

@aelouass.bsky.social

140 followers 1.4K following 39 posts

CS / DS / ML / AI (whatever it is called now) Ph.D. Engineer. Past: INSA-Lyon, LIRIS. Main topics: sequential and temporal data, models and, reasoning. Currently playing with manufacturing data and, interested in RL and Graph DL.

Posts Media Videos Starter Packs

Amine El Ouassouli @aelouass.bsky.social · Jul 17

That it’s time-off o’clock.

Reposted by Amine El Ouassouli

Margot Finn @eicathomefinn.bsky.social · Apr 9

'Bluesky has overtaken its flailing rival X in hosting posts related to new academic research, indicating the platform is fast becoming the go-to place for scholars to share their work.'

X’s dominance ‘over’ as Bluesky becomes new hub for research

Data indicates more scholars turning to alternative social media site to post about their work after Elon Musk’s Twitter takeover

www.timeshighereducation.com

130 4.5K 18K

Amine El Ouassouli @aelouass.bsky.social · Mar 6

13 minutes of wisdom.

“No authorities in science”.

Amen to that.

Csaba Szepesvari @skiandsolve.bsky.social · Mar 6

www.youtube.com/watch?v=9_Pe... An interview with Rich. The humility of Rich is truly inspiring: "There are no authorities in science". I wish people would listen and live by this.

TURING AWARD WINNER Richard S. Sutton in Conversation with Cam Linke | No Authorities in Science

YouTube video by Amii

www.youtube.com

Amine El Ouassouli @aelouass.bsky.social · Feb 18

@jfoerst.bsky.social take on how the community sees the ARC Challenge and how we evaluate models and use benchmarks nowadays is 👌.

#more_science_less_hype (please).

PS: Amazing discussion and good brain food, as usual with MLST.

ImageNet Moment for Reinforcement Learning?

YouTube video by Machine Learning Street Talk

www.youtube.com

1 3

Amine El Ouassouli @aelouass.bsky.social · Jan 31

There is nothing truer than this true statement.

Amine El Ouassouli @aelouass.bsky.social · Jan 30

📍

bsky.app/profile/aelo...

Amine El Ouassouli @aelouass.bsky.social · Jan 16

The more I read and listen to current debates in the field, the more I’m convinced that we have a model evaluation crisis.

Amine El Ouassouli @aelouass.bsky.social · Jan 30

I missed this one when it came out but I can tell that it is one of the most useful piece of research I’ve read in a while.

“GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models”

arxiv.org/html/2410.05...

GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models

arxiv.org

Amine El Ouassouli @aelouass.bsky.social · Jan 28

Seeing “very successful” and “have a somewhat loose relationship with the truth” referring to the same people is what I can’t make sense of …

Amine El Ouassouli @aelouass.bsky.social · Jan 24

We really need better brain-power allocation. The current algorithm is kind of turning crazy.

Ted Underwood @tedunderwood.com · Jan 24

I suspect we’re going to find there is little demand for Operator-like agents that go out on the web and do shopping for you—but huge demand for personal assistants that have read all your email and all your files and can see the 3 tasks you forgot to finish.

Amine El Ouassouli @aelouass.bsky.social · Jan 24

The ToDo list: a revolution.

Amine El Ouassouli @aelouass.bsky.social · Jan 20

Est ce qu'on dirait que "la promotion à outrance des mathématiques est idéologique" ? ça laisse entendre que "l'IA" spécifiquement, à forte dose, est une sciences eugénisante par essence. On peut très bien en faire une interprétation/usage "progressiste" même si ce n'est pas dans l'air du temps.

Amine El Ouassouli @aelouass.bsky.social · Jan 20

À mon avis, ce type d’analyse n’aide absolument pas à se faire une opinion. Le « calcul » (parce que c’est ce que c’est au finale) n’est que le calcul. En faire quelque chose d’idéologique par essence est une sur-interprétation biaisée. L’ « IA » elle même ne porte rien du tout.

Amine El Ouassouli @aelouass.bsky.social · Jan 20

😭

Amine El Ouassouli @aelouass.bsky.social · Jan 16

That’s a very good one 👌🏽

Amine El Ouassouli @aelouass.bsky.social · Jan 16

Best: the most useful research you can do in the current context.
Worst: seems that it is not the main focus for now + maybe the pushbacks.

Amine El Ouassouli @aelouass.bsky.social · Jan 16

May the force be with you!

Amine El Ouassouli @aelouass.bsky.social · Jan 16

Is it an outlier, though?

(and one way of coping for me is to listen to MLST to hear more nuanced, or at sounder, views and opinions + reading)

1 3

Amine El Ouassouli @aelouass.bsky.social · Jan 16

100%

ændra. @aendra.com · Jan 16

Great to see orgs without any prior involvement with the dev community on here very successfully fundraising for fairly nebulous goals while the entire dev community is paying out-of-pocket for tools all y'all are using daily.

Amine El Ouassouli @aelouass.bsky.social · Jan 16

The more I read and listen to current debates in the field, the more I’m convinced that we have a model evaluation crisis.

Amine El Ouassouli @aelouass.bsky.social · Jan 11

I never understood people going to concerts to spend their time there attending through the tiny screens of their phones.

Amine El Ouassouli @aelouass.bsky.social · Jan 8

Basically, IMO, given that all assertions have different degrees of consensus in the population, accurate sequential token prediction may overlap or not with accurate “truth” in the “meaning” or conceptual realm.

Amine El Ouassouli @aelouass.bsky.social · Jan 8

The model may represent truthfully what’s in the dataset, even if it is untruthful. An analogy I often use: you don’t decide if evolution exists by popular vote. The vote tells what the population thinks. Research work even if it is coming from a single individual is more relevant in “truthfulness”

1 7

Amine El Ouassouli @aelouass.bsky.social · Jan 7

Is it just me or are we in an Eliza effect pandemic?

Amine El Ouassouli @aelouass.bsky.social · Jan 6

What is clear for me is that the current hype is not helping the calm development of these methods and collaboration with other fields.

PS: as someone mentioned, cross domain collaboration is key when it comes to ai research. It is hard, but it is key.

Amine El Ouassouli @aelouass.bsky.social · Jan 6

If it is not satisfactory at an epistemological level, it is not always clear at the moment and advances in the field will highlight that later. Is that non-integrity ? I would say no (maybe I’m mistaken).