Machine learner. Building big Bayesian models @microsoft. Views my own. he/him.
Cambridge, England
Joined May 2008
- Tweets536
- Following2K
- Followers7.2K
- Likes931
James Hensman retweeted
Today we bring the latest #DeepSeek distilled models to #Copilot+ PC’s, where they really shine due to the devices’ efficiency and power. This performance is unprecedented. It’s so enabling to put this class of #AI reasoning running on your Copilot+ PC!
blogs.windows.com/windowsdev…
James Hensman retweeted
Excited to share our new work, "TurboAttention", a comprehensive approach for efficient quantized attention in LLMs! We introduce FlashQ, a headwise progressive quantization technique for both KV cache compression and quantized attention execution, and Sparsity-based Softmax Approximation (SAS) to eliminate softmax latency. TurboAttention achieves up to 1.8x speedup in attention, >4.4x KV cache reduction, and up to 2.37x throughput improvement over the FP16 baseline.
Read more: arxiv.org/pdf/2412.08585
#LLM #Quantization #FlashAttention #AI #MachineLearning #Inference #DeepLearning
We're recruiting interns for summer '25. Come and work with me and the team on efficiency in AI systems.
jobs.careers.microsoft.com/g…
James Hensman retweeted
Ever thought about using matrix-exp's or log-det's in large-scale ML (think PDEs/GGN matrices with >1M rows)? I have, and maybe you have, too—but gradients? That's where it gets tricky.
Not anymore! We have a #NeurIPS2024 spotlight: "Gradients of Functions of Large Matrices."🧵
James Hensman retweeted
🔵New paper!🔵 Our latest work on Pyramid Vector Quantization for LLMs achieves state-of-the-art post-training quantization with a Pareto-optimal trade-off between performance, bits per weight, and bits per activation. A thread. 👇 1/15
Our team is growing. Come and join us as part of the MSR AI Residency program to work on efficiency in deep learning.
jobs.careers.microsoft.com/g…
James Hensman retweeted
New preprint out on geographic location encoding using spherical harmonics and sinusoidal representations! 🌐➡️🌏
Work led by @MarcCoru and w/ @rolf_comma_e, Robin Zbinden and @devistuia.
📄Paper: arxiv.org/abs/2310.06743
💻Code: github.com/marccoru/location…
🧵Thread: 👇
1/11
James Hensman retweeted
One week left to apply for a Health Data Science Post Doc with me as part of the @ai4cihub!
bath.ac.uk/jobs/Vacancy.aspx…
There is lots of flexibility in methodology you could pursue and an unbeatable team of collaborators!
James Hensman retweeted
We are hiring a senior researcher in ML for healthcare at MSR Cambridge (UK)! The position is in my team, so if you get it you will work with me (is this a pro or a con? do not answer). Focus is multimodal (~vision-language) models for radiology! Link: jobs.careers.microsoft.com/g…
James Hensman retweeted
Awesome project. Just did an installation video of QuaRot locally. Thanks for the great project.
youtu.be/r7FvhB1WgI8?si=t_4E…
Come and work with us to make Language Models more efficient. A Research Residency at @MSFTResearch and 365 Research in Cambridge, UK.
jobs.careers.microsoft.com/g…
It was such a pleasure to work with @AshkboosSaleh and @max_croci on this method for LLM compression during Saleh's internship last year.
This Post is from an account that no longer exists. Learn more
James Hensman retweeted
We're hiring!🚀
The Alexandria team (microsoft.com/en-us/research…) at @MSFTResearch (Cambridge, UK) is looking for summer #interns. Come work with us on KB-powered LLMs, generative knowledge linking, transformer compression, & other exciting problems
More info: jobs.careers.microsoft.com/g…
Internship opportunity: come and work with me on the future of knowledge bases and efficient language modelling.
jobs.careers.microsoft.com/g…
James Hensman retweeted
State-of-the-art CIFAR-10 results for kernel methods! Now on arXiv:Convolutional Deep Kernel Machines. Link here: arxiv.org/abs/2309.09814. Read on for the abridged version. (1/10)
James Hensman retweeted
The first "Research Focus" blog post of 2024 from @MSFTResearch highlights papers from couple of internship projects that I co-supervised with Siân Lindley and @jameshensman last summer.
Blog post: microsoft.com/en-us/research…
Mixture-of-linear-experts for long-term time series forecasting; Weakly-supervised streaming multilingual speech model with truly zero-shot capability; KBFormer: Diffusion model for structured entity completion; Identifying risks of AI-mediated data access:msft.it/6011imrbT
James Hensman retweeted
A highlight of the #NeurIPS2023 meetup in Cambridge for me was Carl Rasmussen's keynote on "Halting Climate Change". Spoiler: it has nothing to do with AI.
Worth a watch: youtube.com/watch?v=naFaQsFx…
We're looking for a PhD student, apply here:
linkedin.com/jobs/view/37278…
The position is at Lancaster University with @CJ_Nemeth and Paul Fearnhead, co-advised by me from MSR Cambridge.
Lancaster is a super place to study and Stats is a great department. I can't wait to work with you - please apply!