Adam Lee @jimzersML

I like machine learning & web development. adamlee.me California Joined May 2020

Tweets

44
Followers

48
Following

394
Likes

94

M Ganesh Kumar @mgkumar138

9 months ago

Our preprint with @blake__bordelon, @jzavathoneveth, @CPehlevan is now out! We develop a simple reinforcement learning model that recapitulates 3 disparate hippocampal dynamics. Through ablation studies, these representations improve the speed and flexibility of policy learning.

bioRxiv Neuroscience @biorxiv_neursci

9 months ago

0 2 7 6K 3

1 8 26 3K 2

GitHub @github

2 years ago

Your step-by-step guide to building a large language model app, and a list of resources to get you started. ⬇️ github.blog/2023-10-30-the…

13 275 1K 339K 973

Download Image

Alex Xu @alexxubyte

2 years ago

How does Twitter recommend “For You” Timeline in 1.5 seconds? The process involves 5 stages: 🔹 Candidate Sourcing 🔹 Global Filter 🔹 Scoring 🔹 Filtering 🔹 Mixing @elonmusk, I'd love to work with you and your engineers on a book about X architecture. Interested?

20 145 697 113K 312

Download Image

Peyman Milanfar @docmilanfar

3 years ago

There are two kinds of papers. The kind you download and forget to read, and the kind you print and don’t read.

6 22 250 25K 4

Michael Winding @WindingMichael

3 years ago

Check out our new preprint: the complete #connectome of the entire #Drosophila larval brain, with all inputs and all outputs, and a comprehensive analysis of feedforward and feedback pathways across both brain hemispheres #Neuroscience biorxiv.org/content/10.110…

12 179 784 0 99

Download Image

Talmo Pereira 🇧🇷 @talmop

3 years ago

📢 @MurthyLab @shaevitz and I are excited to announce 📢: SLEAP, our multi-animal pose tracking system, is now out in @naturemethods! Paper → nature.com/articles/s4159… Download + docs + data + more → sleap.ai Read on to learn what’s new!

13 169 573 0 73

Download Video

Sergey Levine @svlevine

5 years ago

We've been studying why deep RL is so hard, and we think we have another reason: implicit under-parameterization: arxiv.org/abs/2010.14498 Iteratively training on your own targets is a kind of "self-distillation," and leads to loss of rank -> w/ Aviral Kumar @agarwl_ @its_dibya

3 53 278 0 53

Download Image

François Chollet @fchollet

5 years ago

If you're using Colab and you feel like training your model on GPU is slow, switch to the TPU runtime and tune the "steps_per_execution" parameter in compile(). Seeing a 5-10x speedup is pretty common.

11 132 952 0 219

PyTorch @PyTorch

5 years ago

PreferredNet is moving ChainerRL to the PyTorch ecosystem. With a heavy emphasis on reproducibility, PFRL users have the ability to choose amongst the library's several algorithms and combine them with a multitude of features to produce novel RL systems. medium.com/pytorch/pfrl-a…

2 90 299 0 35

Made With ML @MadeWithML

5 years ago

🏆 Today's trending posts are both on segmentation: - Image Segmentation Using Keras and @weights_biases by @ayushthakur0 madewithml.com/projects/2442/… - SemTorch: library of architectures for segmentation in @PyTorch and can be trained easily with @fastdotai madewithml.com/projects/2388/…

0 33 134 0 34

Download Image

Michael Feathers @mfeathers

5 years ago

Ideally, the next 10 years is the industry internalizing the lessons of Erlang. ferd.ca/a-pipeline-mad…

6 75 159 0 17

Chris Sev @chris__sev

5 years ago

🎊 @hacktoberfest signups are live! October is coming. Time to get our open source on. (and get that sweet tshirt) hacktoberfest.digitalocean.com #DOLove

5 47 167 0 9

Download Image

Daniel Hanley @calamardh

5 years ago

@Buntworthy @Norod78 This is how we created the lasagna cat video, interpolating between both latent representations and generator weights as we retrained the cat model on lasagna:

Lumpy @LumpyTouch

5 years ago

@Buntworthy @Norod78 This is how we created the lasagna cat video, interpolating between both latent representations and generator weights as we retrained the cat model on lasagna:

14 180 777 0 27

Download Video

1 1 6 0 1

PyTorch @PyTorch

5 years ago

The full hands-on tutorials, "Building Recommender Systems with PyTorch", are now available. We show how to build deep learning recommendation system and resolve the associated interpretability, integrity, and privacy challenges. See: youtube.com/playlist?list=…

3 191 823 0 226

Quentin Lhoest 🤗 @lhoestq

5 years ago

🔥After a summer of hard work we are releasing 🤗Datasets v1.0 Started as a way to simplify datasets/metrics access for researchers & teachers And soon became a test bed for efficient & fast data loading/processing :) v1 feats: speed, indexing, multi-modality, reproducibility

6 142 599 0 91

Download Image

Papers with Code @paperswithcode

5 years ago

🔥 Introducing... the ML Reproducibility Challenge 2020! The 4th annual edition now expands to cover papers from 7 major ML conferences: NeurIPS, EMNLP, ACL, ICML, ICLR, CVPR and ECCV. Find more here at Papers With Code: paperswithcode.com/rc2020

0 216 700 0 83

TensorFlow @TensorFlow

5 years ago

🎵🦈 Baby shark doo, doo, doo, doo See how George's Android app uses the SPICE model for pitch detection from #TFHub. The model can convert singing audio into musical notes. Check it out and try it yourself! Post → goo.gle/2EZb0wb Github → goo.gle/3jOjWTF

0 54 171 0 21

Download Video

Andrej Karpathy @karpathy

5 years ago

good quick tutorial on optimizing your PyTorch code ⏲️: youtube.com/watch?v=9mS1fI… quick summary:

27 747 3K 0 1K

Download Image

Hugging Face @huggingface

5 years ago

Want speedy transformers models w/o a GPU?! 🧐 Starting with transformers v3.1.0 your models can now run at the speed of light on commodity CPUs thanks to ONNX Runtime quantization!🚀. Check out our 2nd blog post with ONNX Runtime on the subject! 🔥 medium.com/microsoftazure…

3 167 618 0 91

Ada Yeo @adayeoyh

5 years ago

Poaching an engineer from a friend’s startup after raising funding in Silicon Valley is the equivalent of hitting on your friend’s girlfriend while drunk