Reposted
Reposted
Put this whole paragraph on repeat.
December 4, 2024 at 4:06 PM
Reposted
If you want to keep your knowledge breadth in AI and ML sharp, when a new Lilian Weng blog drops, you read it.
Read "Reward Hacking in Reinforcement Learning" here (lilianweng.github.io/posts/2024-1... ):
Read "Reward Hacking in Reinforcement Learning" here (lilianweng.github.io/posts/2024-1... ):
Reward Hacking in Reinforcement Learning
Reward hacking occurs when a reinforcement learning (RL) agent exploits flaws or ambiguities in the reward function to achieve high rewards, without genuinely learning or completing the intended task....
lilianweng.github.io
December 4, 2024 at 4:26 AM
If you want to keep your knowledge breadth in AI and ML sharp, when a new Lilian Weng blog drops, you read it.
Read "Reward Hacking in Reinforcement Learning" here (lilianweng.github.io/posts/2024-1... ):
Read "Reward Hacking in Reinforcement Learning" here (lilianweng.github.io/posts/2024-1... ):
Reposted
Signal is famous for knowing as little about its users as possible. The app isn’t hoarding metadata, tracking you, or showing you ads.
President @meredithmeredith.bsky.social sees a shift underway as the monoliths of big tech lose popularity and the old economics of Silicon Valley become brittle.
President @meredithmeredith.bsky.social sees a shift underway as the monoliths of big tech lose popularity and the old economics of Silicon Valley become brittle.
With Threats to Encryption Looming, Signal’s Meredith Whittaker Says ‘We’re Not Changing’
At WIRED’s The Big Interview event, the president of the Signal Foundation talked about secure communications as critical infrastructure and the need for a new funding paradigm for tech.
www.wired.com
December 3, 2024 at 8:39 PM
Signal is famous for knowing as little about its users as possible. The app isn’t hoarding metadata, tracking you, or showing you ads.
President @meredithmeredith.bsky.social sees a shift underway as the monoliths of big tech lose popularity and the old economics of Silicon Valley become brittle.
President @meredithmeredith.bsky.social sees a shift underway as the monoliths of big tech lose popularity and the old economics of Silicon Valley become brittle.
Reposted
Ported over from X!
What will a foundation model for the brain look like? 🧠
We argue that it must be able to solve a diverse set of tasks across multiple brain regions and animals.
Check out our NeurIPS paper which introduces a multi-region, multi-animal, multi-task model arxiv.org/abs/2407.14668
What will a foundation model for the brain look like? 🧠
We argue that it must be able to solve a diverse set of tasks across multiple brain regions and animals.
Check out our NeurIPS paper which introduces a multi-region, multi-animal, multi-task model arxiv.org/abs/2407.14668
December 2, 2024 at 10:22 PM
Ported over from X!
What will a foundation model for the brain look like? 🧠
We argue that it must be able to solve a diverse set of tasks across multiple brain regions and animals.
Check out our NeurIPS paper which introduces a multi-region, multi-animal, multi-task model arxiv.org/abs/2407.14668
What will a foundation model for the brain look like? 🧠
We argue that it must be able to solve a diverse set of tasks across multiple brain regions and animals.
Check out our NeurIPS paper which introduces a multi-region, multi-animal, multi-task model arxiv.org/abs/2407.14668
Reposted
Reposted
(1/5) Fun fact: Several classic results in the stat. mech. of learning can be derived in a couple lines of simple algebra!
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
Simplified derivations for high-dimensional convex learning problems
Statistical physics provides tools for analyzing high-dimensional problems in machine learning and theoretical neuroscience. These calculations, particularly those using the replica method, often invo...
arxiv.org
December 3, 2024 at 7:34 PM
(1/5) Fun fact: Several classic results in the stat. mech. of learning can be derived in a couple lines of simple algebra!
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
In this paper with Haim Sompolinsky, we simplify and unify derivations for high-dimensional convex learning problems using a bipartite cavity method.
arxiv.org/abs/2412.01110
Reposted
🎺 Here comes the official 2024 NeurIPS paper browser:
- browse all NeurIPS papers in a visual way
- select clusters of interest and get cluster summary
- ZOOOOM in
- filter by human assigned keywords
- filter by substring (authors, titles)
neurips2024.vizhub.ai
#neurips by IBM Research Cambridge
- browse all NeurIPS papers in a visual way
- select clusters of interest and get cluster summary
- ZOOOOM in
- filter by human assigned keywords
- filter by substring (authors, titles)
neurips2024.vizhub.ai
#neurips by IBM Research Cambridge
December 3, 2024 at 5:01 PM
🎺 Here comes the official 2024 NeurIPS paper browser:
- browse all NeurIPS papers in a visual way
- select clusters of interest and get cluster summary
- ZOOOOM in
- filter by human assigned keywords
- filter by substring (authors, titles)
neurips2024.vizhub.ai
#neurips by IBM Research Cambridge
- browse all NeurIPS papers in a visual way
- select clusters of interest and get cluster summary
- ZOOOOM in
- filter by human assigned keywords
- filter by substring (authors, titles)
neurips2024.vizhub.ai
#neurips by IBM Research Cambridge
Reposted
"We cannot continue to participate in the social network feed of a man who proclaims the death of the media and therefore of journalists... X is a platform that no longer serves the public interest at all, but the special ideological and financial interests of its owner and his political allies"
European Federation of Journalists to stop posting content on X
The EFJ represents over 295,000 journalists in 44 countries.
www.thelondoneconomic.com
December 3, 2024 at 6:22 AM
"We cannot continue to participate in the social network feed of a man who proclaims the death of the media and therefore of journalists... X is a platform that no longer serves the public interest at all, but the special ideological and financial interests of its owner and his political allies"
Reposted
So proud of the great work happening in Minneapolis, which shows that boosting housing supply helps make homes more affordable.
December 3, 2024 at 3:02 PM
So proud of the great work happening in Minneapolis, which shows that boosting housing supply helps make homes more affordable.