We use pre-trained vision-language models to improve video classification, achieving SOTA performance on Kinetics-400 and surpassing previous methods by 20-50% on five popular video datasets. #AAAI2023 Github: https://
github.com/whwu95/Text4Vis
MACHINE LEARNING
-
Vision-Language Models Achieve SOTA Video Classification Performance
By
–
-
LandingLens: Computer Vision Projects Now Completed in Days
By
–
3/But building and deploying ML systems is hard. After years of work, I’m thrilled to finally offer LandingLens to everyone. Computer vision projects that used to take me a year are now done in days/weeks on LandingLens. Please try it out and also help me spread the word!
-
Democratizing AI: Custom ML Models for Everyone’s Data
By
–
2/We’re on a mission to democratize access to AI creation, so everyone can easily train and deploy custom ML models on their own data. Every business and every person now has their own data, so using only models that someone else had trained on their data is not good enough.
-
Landing AI Releases Free Computer Vision Tool for Image Labeling and Model Deployment
By
–
1/Landing AI is releasing our computer vision tool for anyone to use! You can label images, train a model, and deploy your model to production in minutes. Our data-centric tool then helps you improve your model quickly. Start for free at https://
landing.ai -
Groq Welcomes Matanya Loewenthal to Machine Learning Team
By
–
We're thrilled to include Matanya Loewenthal on our team of creative problem solvers, helping us build the next generation of Machine Learning Systems! Join us: https://t.co/fuR1jBakSQ#WhyGroq #hiring pic.twitter.com/69wL1HR7Ri
— Groq Inc (@GroqInc) 22 février 2023We're thrilled to include Matanya Loewenthal on our team of creative problem solvers, helping us build the next generation of Machine Learning Systems! Join us: http://
groq.com/careers
#WhyGroq #hiring -
Daily Python, Machine Learning and Language Models Tutorials
By
–
Every day, I share tutorials and simplify complex topics around Python, Machine Learning & Language Models. Follow me → @Sumanth_077 to ensure you don't miss that. Like/RT the first tweet to support my work and help this reach more people.
-
Open Source Solution Replicates ChatGPT Training Process Efficiently
By
–
Open Source Solution Replicates ChatGPT Training Process! Ready To Go With Only 1.6GB GPU Memory And Gives You 7.73 Times Faster Training! https://
syncedreview.com/2023/02/22/ope
n-source-solution-replicates-chatgpt-training-process-ready-to-go-with-only-1-6gb-gpu-memory-and-gives-you-7-73-times-faster-training/
… -
OpenAI’s Foundry: GPT DV 32k model could generate 8,000+ words per prompt
By
–
OpenAI's "Foundry" product brief: DV 32k model instance could mean this GPT model is capable of producing 8,000+ words per prompt
-
AI Restores Low-Quality Images Through Advanced Technology
By
–
Impressive restoration of memories by AI Even very low-quality inputs can be restored Credit: GFPGAN
#innovation #artificialintelligence #machinelearning -
German AI startup Aleph Alpha claims 300-billion parameter LLM Luminous-World
By
–
German AI startup @Aleph__Alpha claims it will have a 300-billion parameter Large Language Model "Luminous-World" trained this year for “highly complex and critical applications” (GPT-3 has 175 billion parameters)