We really enjoyed meeting many brilliant ML researchers and model builders at the Cerebras booth last week at NeurIPS. Some topics of discussion: * Unlocking Sparsity for Training Efficiency
* Optimizing Training for Inference Efficiency
* Multilingual Models
* Training large
MACHINE LEARNING
-
Cerebras Booth Highlights: Sparsity, Inference Optimization, Multilingual Models
By
–
-
Fastest BS=1 Inference with PyTorch Speculative Decoding
By
–
I think this breezes past as the fastest bs=1 inference I know.
With plain and simple PyTorch code.
Speculative Decoding isn't even applied, and will be purely additive… -
Software Portability Myths and Machine Learning Progress Implications
By
–
Watch @fraser_mince present “The Grand Illusion: The Myth of Software Portability and Implications for ML Progress” at #NeurIPS2023
— Cohere (@cohere) 18 décembre 2023
Work with @dzungdinhh @jonas_kg @ProfNeilT @sarahookr at @CohereForAI and @mit_ide
Read more: https://t.co/JWjzYYFFa0 pic.twitter.com/qvvobaWLSVWatch @fraser_mince present “The Grand Illusion: The Myth of Software Portability and Implications for ML Progress” at #NeurIPS2023 Work with @dzungdinhh @jonas_kg @ProfNeilT @sarahookr at @CohereForAI and @mit_ide Read more:
-
Multiscale Hamiltonian Monte Carlo with Delayed Rejection
By
–
Bob Carpenter: Multiscale Generalized Hamiltonian Monte Carlo with Delayed Rejection https://
bit.ly/3uNE57o #AI #MachineLearning #DeepLearning #LLMs #DataScience -
Stanford AI Leaders Forecast Top 2024 Artificial Intelligence Stories
By
–
AI forecasts for 2024: @StanfordHAI
’s leading minds predict the biggest stories for next year in artificial intelligence. @erikbryn @landay @Rbaltman @drfeifei @gewang @kingjen -
Understanding AI Theory: How Much Knowledge Is Necessary?
By
–
Is the theory important for leveraging "AI technologies"? How much should one understand those technologies, for example how the GPT models, mixture of experts, diffusion networks… work? Some clarifications: Some parts are useful (which? Let me know in the thread!)
-
Machine Learning Predictions: Humility Over Overconfidence
By
–
Using Machine Learning for predictive modeling is a humbling experience. You realize how hard it is even when you have the best data and algorithms. And it makes you more circumspect of all the pundits and “experts” who overconfidently shill their predictions as facts.
-
AI detects feline pain with 77% accuracy via facial recognition
By
–
AI reads cat pain? New study shows 77% accuracy in detecting feline discomfort by analyzing facial expressions! Could this lead to pain-free future for our furry friends?
-
Comprehensive Guide to Contour Detection Using OpenCV
By
–
🎥New Video Alert🎥https://t.co/XpN3Lu7mBp
— Satya Mallick (@LearnOpenCV) 18 décembre 2023
Check out our comprehensive guide as we delve into Contour Detection using OpenCV, walk you through the basics, and explore some of its applications.#AI #contourdetection #computervision #deeplearning #opencv #learnopencv pic.twitter.com/mOrM3HR9MWNew Alert https://
youtube.com/watch?v=FKIn_U
IsXy4
… Check out our comprehensive guide as we delve into Contour Detection using OpenCV, walk you through the basics, and explore some of its applications. #AI #contourdetection #computervision #deeplearning #opencv #learnopencv -
UNet Architecture Acceleration for Diffusion Models Training-Free
By
–
Check our recent work on diffusion model acceleration from the perspective of UNet architecture! Our approach can easily be combined with various diffusion model-based tasks (text-to-image, personalized generation, video generation, etc.) to achieve training-free acceleration.