David Low @davidlowjw

Director of Advanced Analytics @ Singtel davidlowjw.github.io Singapore Joined September 2010

Tweets

273
Followers

148
Following

311
Likes

173

Andrej Karpathy @karpathy

2 years ago

Promising. Everyone should hope that we can throw away tokenization in LLMs. Doing so naively creates (byte-level) sequences that are too long, so the devil is in the details. Tokenization means that LLMs are not actually fully end-to-end. There is a whole separate stage with…

AK @_akhaliq

2 years ago

15 216 1K 1.5M 582

Download Image

86 585 4K 1.5M 2K

Jim Fan @DrJimFan

3 years ago

Why does ChatGPT work so well? Is it “just scaling up GPT-3” under the hood? In this 🧵, let’s discuss the “Instruct” paradigm, its deep technical insights, and a big implication: “prompt engineering” as we know it may likely disappear soon:👇

48 480 2K 0 1K

Download Image

Xavier Bresson @xbresson

4 years ago

I recently spoke with a journalist about applications of Graph Neural Networks. I made a few slides to facilitate the discussion. I share them here -- hopefully they can be useful :) dropbox.com/s/u82gcg3aath3…

1 71 284 0 65

hardmaru @hardmaru

4 years ago

How many can you recognize? 👶🏻 (From @TheInsaneApp)

31 113 802 0 50

Download Image

Xavier Bresson @xbresson

4 years ago

I have a postdoctoral opening for a project GraphNNs+Ophthalmology. The position is available asap. The duration is one year, renewable another year on mutual agreement. Given the current restrictions, priority will be given to Singaporeans/PRs.

1 8 33 0 1

Natasha Jaques @natashajaques

4 years ago

I couldn't resist. With contributions from @maxhkw

61 1K 6K 0 397

Download Image

Xavier Bresson @xbresson

5 years ago

My main talks on Graph Neural Networks in 2020 1. Introduction to GNNs youtube.com/watch?v=Iiv9R6… 2. Recent developments in GNNs youtube.com/watch?v=M60hux… 3. Benchmarking GNNs youtube.com/watch?v=tuChBS… Hope they can be useful. Happy new year to everyone !

8 370 2K 0 559

Download Image

Xavier Bresson @xbresson

5 years ago

Sharing my lecture slides on "Recent Developments of Graph Network Architectures" from my deep learning course. It is a review of some exciting works on GNNs published in 2019-2020. #feelthelearn rb.gy/quo3n6

11 246 1K 0 355

Download Image

Xavier Bresson @xbresson

5 years ago

Hi #Singapore! Do you want to learn #AI and #DeepLearning with an #expert? I will #teach a 3-day #course (lectures and coding exercises) on Dec 7-9 at One-North. Register here: rb.gy/xuvh2a @SkillsFutureSG @SmartNationSG @SGInnovate

0 1 13 0 2

Download Image

Chaitanya K. Joshi @chaitjo

5 years ago

"Multi-Head Attention: Collaborate Instead of Concatenate" -@jb_cordonnier @loukasa_tweet M. Jaggi (EPFL). Multi-head attention is critical to train Transformers, but most attention heads can later be pruned/are redundant. This paper asks why! 🧐 Paper: arxiv.org/abs/2006.16362

1 48 216 0 41

Download Image

Xavier Bresson @xbresson

5 years ago

.@ylecun @OriolVinyalsML @wellingmax @PeterWBattaglia @69alodi S. Jegelka S. Osher and I are organizing an exciting workshop on the intersection of deep learning and combinatorial optimization at @ipam_ucla on Feb 22-26 2021. ipam.ucla.edu/programs/works…

1 38 175 0 21

Download Image

hardmaru @hardmaru

5 years ago

Your state-of-the-art method is only state-of-the-art if other people can actually use it.

10 89 722 0 6

Denny Britz @dennybritz

5 years ago

I built a library and API to analyze and visualize the AI/ML citation graph. It comes with notebooks that you can run locally or in Google Colab: github.com/dennybritz/pap… What is this useful for? Here's an example (1/3)

7 82 341 0 86

Jake VanderPlas @jakevdp

5 years ago

The frequency of random seeds between 0 and 1000 on github (data from grep.app)

69 375 2K 0 65

Download Image

Richard Socher @RichardSocher

6 years ago

Introducing ProGen, a large language model trained on 280 million protein sequences that can generate viable proteins based on user specifications. A step towards AI & #nlpproc helping cure disease and clean our planet. Paper: biorxiv.org/content/10.110… Blog: blog.einstein.ai/progen

10 102 336 0 31

Download Image

Xavier Bresson @xbresson

6 years ago

New paper on benchmarking graph neural networks w/ @vijaypradwi @chaitjo T. Laurent and Y. Bengio arxiv.org/pdf/2003.00982… Our goal was to identify trends and good building blocks for GNNs.

8 126 455 0 93

Download Image

Chaitanya K. Joshi @chaitjo

6 years ago

Excited to share a blog post on the connection between #Transformers for NLP and #GraphNeuralNetworks (GNNs or GCNs). graphdeeplearning.github.io/post/transform…

16 240 826 0 283

Download Image

Olivier Grisel @ogrisel

6 years ago

Plotly's interactive parallel coordinates plot is really nice to build some insights on the impact of hyper-parameters on the generalization performance.

9 107 500 0 92

Download Video

evolvingstuff @evolvingstuff

6 years ago

Transformers as Soft Reasoners over Language "we explore whether transformers can similarly learn to reason (or emulate reasoning), but using rules expressed in language, thus bypassing a formal representation." arxiv.org/abs/2002.05867 Datasets and demo: rule-reasoning.apps.allenai.org

7 48 179 0 29

Download Image

OpenAI @OpenAI

6 years ago

We've trained an AI system to solve the Rubik's Cube with a human-like robot hand. This is an unprecedented level of dexterity for a robot, and is hard even for humans to do. The system trains in an imperfect simulation and quickly adapts to reality: openai.com/blog/solving-r…