Really cool playground for all things related to preprocessing Glad to see this getting the attention it deserves!
MACHINE LEARNING
-
Differential Privacy Composition Cost in Multi-Step ML
By
–
When applying #DifferentialPrivacy to multi-step #ML algorithms, there is a privacy loss that scales with each step, known as the cost of composition. Learn about a new paradigm that avoids this composition cost for significant improvements in utility. ↓
https://
goo.gle/44NMl3z -
Streamlining Object Tracking with Deep SORT and Torchvision
By
–
🎥New Videohttps://t.co/YoZgWMG8LL
— Satya Mallick (@LearnOpenCV) 11 septembre 2023
Check out our latest video where we delve into streamlining object tracking using Deep SORT and Torchvision detectors.#computervision #ai #deeplearning #torchvision pic.twitter.com/eNKqYPPlT1New https://
youtube.com/watch?v=GkZRKa
QZ_ys
… Check out our latest video where we delve into streamlining object tracking using Deep SORT and Torchvision detectors. #computervision #ai #deeplearning #torchvision -
Rerank Endpoint: Semantic Search Simplified Single Line Code
By
–
Did you know that the Rerank endpoint needs just a single line of code to implement? The Rerank endpoint acts as the last stage re-ranker of a search flow, giving you a performance boost from semantic search minus the complexities. See how it works using Wikipedia Search as an
-
Learning AI Skills: Relevance vs. Investment in Rapid Evolution
By
–
You think so! And what about learning? As in, learning about AI, how to train the models or productive them? Do you feel like this is relevant vs. investing in AI, or is this also evolving so fast that the acquired skills become obsolete, too?
-
PhD vs Startup vs Corporate Job in AI: Career Path Comparison
By
–
It's definitely the ideal case! But you can do that while doing a PhD. Masters or a regular job, too! Would you recommend going straight for projects vs. studies or professional work? Let's say you can 100% land a startup or regular company AI-related role, do a PhD or
-
GPT-4 Falcon Llama2 Tokenizer Comparison Study Results
By
–
And our answers are out! Running on 1B tokens from the web (filtered and mostly in English as details in https://
huggingface.co/papers/2306.01
116
…) we got – GPT4 tokenizer (100k vocab) gives you 0.997B tokens – Falcon tokenizer (64k vocab) gives you ~5% more tokens (1.04B)
– Llama2 tokenizer -
AI’s New Frontier: Understanding Cause and Effect
By
–
Has AI just crossed a mind-bending frontier? Can machines now understand Cause and Effect?
-
Affective Image Captioning with Contextual Cues Analysis
By
–
Hi @acii_conf
! Interested in emotions and context? Vera Yang presents a new affective image captioning method to explore the effect of adding or subtracting contextual cues! 1:15pm in MR (LLM session) #acii2023 -
AI Frontier: Can Machines Understand Cause and Effect?
By
–
Has AI just crossed a mind-bending frontier? Can machines now understand Cause and Effect? I've frequently argued that a fundamental difference between humans and Artificial Intelligence is our innate ability to understand causal reasoning, …