Ian (@foretdepins) header
Ian (@foretdepins) profile picture

Life is family, statistics, machine learning, R, Python, running and rugby. Uni. of Southampton MSc Data Science grad, sports research, analyst in aerospace.

Hampshire
Joined February 2019
Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)Photo by Ian (@foretdepins)
Excited to be presenting our paper “Seq2Event: learning the language of soccer using transformer-based match event prediction” (available: eprints.soton.ac.uk/458099/ ) at #KDD2022 @ryanbeal95 @proftjn @kdd_news
1
2
1
6
Genuinely looking forward to this Rugby World Cup, set to possibly be the most exciting ever, especially in the pool stages!! #RWC2023 🇫🇷🏉
4
163
Keynote: “My Enemy’s Enemy is my Friend: How Statisticians and Machine Learners Should Work Together to Build a Better World” by @MihaelaVDS. A solid reflection on how the two fields should be used together to better solve real world problems. #RSS2023Conf @RSSAnnualConf – at Harrogate, England
1
110
Also, intrigued by her lab’s current work on Casual Deep Learning, attempting to go beyond association in ML. @MihaelaVDS #RSS2023Conf @RSSAnnualConf – at Harrogate, England
78
Delighted to be settling in to the Royal Statistical Society international conference 2023 in Harrogate. Great location, great talks so far! @RSSAnnualConf #RSS2023conf – at Harrogate, England
1
84
This is huge: Llama-v2 is open source, with a license that authorizes commercial use! This is going to change the landscape of the LLM market. Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers Pretrained and fine-tuned models are available with 7B, 13B and 70B parameters. Llama-2 website: ai.meta.com/llama/ Llama-2 paper: ai.meta.com/research/publica… A number of personalities from industry and academia have endorsed our open source approach: about.fb.com/news/2023/07/ll…
379
3,327
751
14,805
4,294,659
Like, I don’t want to ruin the joke by explaining it but the reason this ended up so good and funny is that everyone played along with my goofiness and was in on the joke from the beginning
Hey, I’m trying to improve this unsupervised model to correctly label users. Looking for an intern to improve it. Here’s a picture of the current clusters. If anyone has ideas, feel free to respond to this tweet with a gist of your implementation. Winner gets a lollipop.
2
4
41
Been thinking about @visakanv's "good reply game", aka the idea that good-faith conversation on the internet relies on making OP look good by playing along with them instead of arguing, like in improv, and honestly this makes so much sense about what good communities get right.
10
23
2
223
30 years ago: Transformers with linearized self-attention in NECO 1992, equivalent to fast weight programmers (apart from normalization), separating storage and control. Key/value was called FROM/TO. The attention terminology was introduced at ICANN 1993 sferics.idsia.ch/pub/juergen…
35
154
43
1,269
Hey Neptune. Did you ring? 👋 Webb’s latest image is the clearest look at Neptune's rings in 30+ years, and our first time seeing them in infrared light. Take in Webb's ghostly, ethereal views of the planet and its dust bands, rings and moons: go.nasa.gov/3RXxoGq #IAC2022
420
9,744
1,840
38,259
Fascinating and v.busy press event by James Webb Telescope team #IAC2022 @iafastro @NASAWebb
We are greatly saddened by the news today of the death of Her Majesty the Queen. The University flag will be flown at half-mast as a mark of our respect during the period of formal national mourning.
1
38
160
Ian retweeted
Noteworthy that all the dissertations that received awards at #kdd2022 are about Deep Learning for Graphs.
2
9
2
77
#KDD2022 closing ceremony. It’s been great, an experience I will never forget! Amazing people, amazing talks. @kdd_news
4
“B3ER: BERT-based early-exit using expert routing” presented by Ankith MS from @AmazonScience #KDD2022 @kdd_news
1
3
“Beyond Point Prediction” presented by @andeezymcd, interesting work on deep learning for extreme value spatiotemporal domains. #KDD2022 @kdd_news
2
Attending hands-on workshop “Toolkit for Anomaly Detection” by Dr. Dhaval Patel #KDD2022 @IBMResearch @kdd_news
2
KDD innovation award winner Huan Liu @liuhuan Sharing that his journey started at KDD95. #KDD2022 @kdd_news
2
5
Opening ceremony at #KDD2022! @kdd_news 🎉
3
Attending #DeepSpatial22 at #KDD2022, @StefanoErmon presenting “Monitoring and Accelerating Sustainable Development with AI”.
1
1
Satellite imagery is becoming ever more advanced, but a key problem is a lack of labelled data. But georgraphy-aware self-supervised learning techniques can help to overcome this.