Yusaku Sako

_

[Andrej Karpathy @karpathy]

How to become expert at thing:
1 iteratively take on concrete projects and accomplish them depth wise, learning “on demand” (ie don’t learn bottom up breadth wise)
2 teach/summarize everything you learn in your own words
3 only compare yourself to younger you, never to others

[Yusaku Sako @u39kun]

@smly Wow! What's all the data for?

[〈 Berger | Dillon 〉 @InertialObservr]

A perfectly square grid

(inspired by @DrGBuckingham)

[Miles Cranmer @MilesCranmer]

Very excited to share our new paper "Discovering Symbolic Models from Deep Learning with Inductive Biases"!

We describe an approach to convert a deep model into an equivalent symbolic equation.

Blog/code: https://t.co/k7cWi6RpEj
Paper: https://t.co/xEW0hL7ArT

Thread👇
1/n

[Yusaku Sako @u39kun]

@tunguz @nvidia @rapidsai @Twitter Congrats @tunguz and the @rapidsai team!

[Yusaku Sako @u39kun]

@tunguz 😂

[Yusaku Sako @u39kun]

@jm_alexia That's amazing. Congrats, @jm_alexia!

[Yusaku Sako @u39kun]

Identifying bottlenecks in Python code (and iterating on improvements) is *ridiculously easy* on Jupyter Notebook with %prun and %%prun.
I wish I knew these magic commands sooner, like years ago!
https://t.co/jiBoQmXLQX

#Python @ProjectJupyter #DataScience
#MachineLearning

[Yusaku Sako @u39kun]

@dennybritz @Cometml Have you tried Sacred + Omniboard combo as a free, open source solution, keep-all-your-metadata-local solution to tracking ML experiments?

[Hamel Husain @HamelHusain]

Recently, @jeremyphoward, @fastdotai and I launched fastpages. This post goes behind the scenes to explain the magic that makes it work: @github Actions. We show you additional examples of how this can give you new superpowers for ML & Data Science: https://t.co/sqxlgtWlRs

[Quoc Le @quocleix]

New paper: Towards a Human-like Open-Domain Chatbot. Key takeaways:

1. "Perplexity is all a chatbot needs" ;)
2. We're getting closer to a high-quality chatbot that can chat about anything

Paper: https://t.co/r09PQ9ldea
Blog: https://t.co/QgDnNJ2XiP

[Yusaku Sako @u39kun]

@jm_alexia Looking forward to it! It is great that you keep publishing interesting work on GANs. You are inspirational to independent researchers (um, wannabes) like me!

[Yusaku Sako @u39kun]

@alexeev_eu @jm_alexia @jonathanfly @simon_niklaus Entering the matrix

[Yusaku Sako @u39kun]

@hardmaru I wrote a blog post on this if anyone is interested: https://t.co/KWdwsKp4Ms

[Yusaku Sako @u39kun]

Man, @lexfridman's podcast is pure awesomeness. Waiting for one with @donalddhoffman!!! Please?

[Yusaku Sako @u39kun]

Thanks for tweeting my article, @tunguz. We will not let cheaters destroy the great @kaggle community that we have!

[Bojan Tunguz @tunguz]

Another great take on the latest scandal:

Inevitable downfall of a Kaggle Grandmaster who tried to win at any cost by @u39kun https://t.co/yeFeoEJHSA

[hardmaru @hardmaru]

Hochreiter & Schmidhuber 🔥🔥

Two decades after the LSTM paper

#NeurIPS2019

[Yusaku Sako @u39kun]

https://t.co/xFketisA7m

A new #Kaggle competition for detecting "deepfake" videos just launched.
Prize purse: $1,000,000

#MachineLearning #AI

[Yusaku Sako @u39kun]

This is awesome. I have been wanting something like this that allows rapid notebook-first development while not compromising on modularity/testing/version control/packaging/distribution. Thanks @GuggerSylvain @jeremyphoward @fastdotai , will give it a shot!

[Jeremy Howard @jeremyphoward]

I've been working with @GuggerSylvain for a long time to create something that combines the best of @ProjectJupyter Notebooks with the best of traditional software development approaches.

It's called nbdev. We're releasing it today as open source. https://t.co/XNKFpfuigZ

1/