The ROI is too long term for all private companies (except a handful like Google, Meta, Microsoft, or IBM)
It might take 5, 10, or 20+ years (as it did with deep learning)
@ylecun
-
AI ROI Timeline: Long-Term Investment Challenge for Private Companies
By
–
-
Government Research Funding ROI Critical for AI Innovation
By
–
In this piece published in The Hill, fellow Turing laureate David Patterson points out that the return on investment of government-funded research in universities is gigantic.
Cutting the budgets of NSF and NIH is economic suicide. -
World Models: Action Conditioning and Representation Learning
By
–
– conditioning on actions is what makes it a world model
– planning action sequences is what makes it useful.
– predicting in representation space trained in a completely self-supervised, task-independent manner is what makes it complicated.
– training the encoder and predictor -
Meta Researcher Balances Industry and Academic Roles at NYU
By
–
This was long before I worked at FB/Meta.
But yes. I share my time between Meta and NYU. -
Continuous vs Discrete Prediction Space in Modern AI Models
By
–
This model makes predictions in continuous representation space.
An LLM makes predictions in discrete input space.
Prediction in continuous representation space is done with a (non-generative) Joint Embedding Predictive Architecture, i.e what I've been advocating for about 5 -
Learning Abstract Representation Spaces for Improved Model Prediction
By
–
I said it was impossible to predict all the details at the pixel level.
I said the solution was to learn an abstract representation space within which prediction could more easily be learned.
Which is *precisely* what this model does. -
Only Profitable Companies Can Fund Meaningful Research Labs
By
–
You're absolutely right. Only large and profitable companies can afford to do actual research.
All the historically impactful industry labs (AT&T Bell Labs, IBM Research, Xerox PARC, MSR, etc) were with companies that didn't have to worry about their survival.
They stopped -
Research to Product Transfer Requires Trust Between Teams
By
–
You put your finger on one reason why the transfer from research to products is a difficult struggle of every day.
It requires trusting relationships between research and development teams.
The "annoyance" you mention is one reason why so much awesome research spends so much time -
Foundational Research Impact on Modern Engineering Teams
By
–
Some of the stuff his engineering teams are building today would not exist without some of the research I published 30 years ago.
-
Scientist collaboration and recruitment in machine learning research
By
–
The scientist in question (let's call him X) wasn't in my lab.
He was hanging around my lab hoping to collaborate with me (I wasn't funding him), and sat in my class, hoping to learn about ML.
Koray was also in my class as a master student.
X figured Koray was good and brought