the « internet data lottery »: improvements in AI/ML coming from the path of largest web data availability rather than most scientifically sound direction modalities (text) win the lottery when they happen to be the widest available on the web even if sub-optimal for some tasks
MACHINE LEARNING
-
Essential AI Experts to Follow Curated List
By
–
Here’s the list of AI folks to follow I’ve been curating for the past month. Let me know if you have any to add https://
x.com/i/lists/158543
0245762441216
… -
Explaining Model Parameters in Machine Learning
By
–
for those of you wondering, what is a parameter? It’s a configuration variable that is internal to the model and whose value can be estimated from the given data. They are required by the model when making predictions. The values define the skill of the model on your problem.
-
Stanford Microsoft Method Enables Natural Language Model Bug Fixes
By
–
Talking to Models: Stanford U & Microsoft Method Enables Developers to Correct Model Bugs via Natural Language Patches https://
syncedreview.com/2022/11/21/tal
king-to-models-stanford-u-microsoft-method-enables-developers-to-correct-model-bugs-via-natural-language-patches/
… -
Data Infrastructure and GPU Acceleration for AI ML Workloads
By
–
With data volumes rapidly increasing, companies need to equip data scientists with storage and data management, GPU acceleration, and tooling to harness #AI and #ml. Experts from Domino, @netapp
, and @nvidia met at #vmwarexplore in Barcelona to discuss: https://
domino.buzz/3OwI12c -
Hierarchical Living Systems: P≠NP and Evolutionary Constraints
By
–
Living systems have hierarchical structure because P ≠ NP, and evolution can only try exponential combinations of a small number of things at a time.
-
Train and Deploy Custom DreamBooth Models on Replicate
By
–
Train and deploy DreamBooth models on Replicate. With just a handful of images and a single API call, you can train your own custom Stable Diffusion, publish it to Replicate, and run predictions on it in the cloud.
-
ML Study Analyzes Gender and Diversity in Television Content
By
–
In partnership with @GeenaDavisOrg & @USC
, we used ML to study perceived gender expression, skin tone, and age representation in TV shows over the past 12 years. We analyzed 440 hours of footage to understand trends in screen & speaking time. Learn more ↓ -
Correction: Chain of Thought paper accepted at NeurIPS
By
–
Oops i had a typo, CoT is in NeurIPS not ICLR