How to Fine-Tune Vision Models with SGD Kumar et al.: https://
arxiv.org/abs/2211.09359 #ArtificialIntelligence #DeepLearning #MachineLearning
RESEARCH
-
Fine-Tuning Vision Models with SGD Optimization
By
–
-
Professional Email Etiquette for Researchers and Tech Community
By
–
a) Don't email people who ask not to be
b) Use common sense (e.g., send 1 email not 18, don't follow up a million times, don't try to track them)
c) Try to *genuinely* engage with their work (don't just namedrop a paper)
d) Prepare yourself to be totally ignored. Its not personal -
Evolution of Object Detection Methods: From Sliding Windows to YOLO
By
–
Traditionally it was with a sliding window approach using an object recognizer passed over the image at different scales. Now some of the fastest methods like YOLO2 or 3 jointly detect bounding boxes and classify at the same time. (I may be a year or 2 out of date.)
-
Meta AI Stanford Tübingen Beat Neural Scaling Laws Data Pruning
By
–
NeurIPS 2022 | Meta AI, Stanford & Tübingen U Beat Neural Scaling Laws via Data Pruning https://
syncedreview.com/2022/11/28/neu
rips-2022-meta-ai-stanford-tubingen-u-beat-neural-scaling-laws-via-data-pruning/
… -
AI Models Will Soon Outperform Humans at Comedy
By
–
Models can’t do comedy reliably yet, but very soon, they’ll be better than us…
-
Andromeda AI Supercomputer: 13.5M Cores, Exascale Computing
By
–
@ykilcher recently spoke about Andromeda, our AI Supercomputer that consists of 13.5M cores, 1 exaFLOP of AI compute, and near-perfect linear scaling. Check out the episode here: https://
hubs.li/Q01tnsyt0 #AI #ML -
GPT-3 Model Improvements: Early Testing Results
By
–
I've been playing around with the new GPT-3 model for a few minutes, and it's already clearly better than the GPT-3 I'm used to using. Will continue to test it out and report findings.
-
Artificial Intelligence Benefits and Drawbacks Explored
By
–
Artificial intelligence is both Yin and Yang https://
dataconomy.com/2022/11/benefi
ts-and-drawbacks-of-artificial-intelligence/
… @DataconomyMedia #ArtificialIntelligence #MachineLearning #Robots #BigData #Analytics #AI #100DaysofCode #IoT #serverless #DEVCommunity #womenwhocode #DigitalTransformation #DataScience #DeepLearning #AIEthics -
DeepMind LLM Research at NeurIPS Discussion and Recruitment
By
–
At NeurIPS this week, reach out if you want to discuss our work around LLM at @DeepMind (Chinchilla, Retro, Flamingo), and if you're interested in working with us!
-
AutoML Pros, Cons, and Misconceptions Revealed
By
–
Automated ML can do a lot – but it can also be a solution looking for a problem. When is it useful or not? In @elleng_glover
's in-depth article for @BuiltIn
, Domino’s Kjell Carlsson reveals the pros, cons, and misconceptions about #AutoML: