Distill

Supporting Clarity in Machine Learning

[Distill @distillpub]

These papers were submitted to Distill prior to it going on hiatus, and are likely among the last papers Distill will publish in the foreseeable future.

[Distill @distillpub]

Today Distill published two expository papers on Graph Neural Networks, listed below.

[Distill @distillpub]

A Gentle Introduction to Graph Neural Networks by @beangoben, @emilyrreif, @adamrpearce, and Alex Wiltschko.

https://t.co/camcvp1cMS

[Distill @distillpub]

Distill will be taking a one year hiatus starting today, which may be extended indefinitely. Our editorial team has written some reflections on what we've learned over the past five years, and the factors that led to this decision.

https://t.co/E31QPnPbMt

[Distill @distillpub]

Adversarial Reprogramming of Neural Cellular Automata — A new Distill article by @RandazzoEttore, @zzznah, @eyvindn, and @drmichaellevin. This is the fourth article in the Differentiable Self-organizing Systems thread.

https://t.co/wMEqtFx4Kh

[Chelsea Sierra Voss @csvoss]

New lenses – tools for seeing more detail more clearly –are essential for any scientific endeavor, and this new Circuits article by @mpetrov et al. provides us with an incredibly crisp example of why.

[Leon Overweel @layon_overwhale]

For today's Dynamically Typed, I finally sat down to read (and summarize!) three recent @distillpub articles, on multimodal neurons, branch specialization, and weight banding.

I'll highlight each one in this thread; all three are also here: https://t.co/AUOsVbZmg7 (1/4)

[Distill @distillpub]

Weight Banding — A new Distill article by
@mpetrov, @csvoss, @ludwigschubert, @nickcammarata, @gabeeegoooh, and @ch402. This is the ninth article in the Circuits thread.

https://t.co/2BZA3UiFsA

[Distill @distillpub]

Branch Specialization — A new Distill article by @csvoss, @gabeeegoooh, @nickcammarata, @mpetrov, @ludwigschubert, and @ch402. This is the eighth article in the Circuits thread.

https://t.co/GJYvF179Km

[Chelsea Sierra Voss @csvoss]

Do neural networks have any emergent structures that are larger-scale than circuits?

In our new Circuits article, we investigate the tendency of neural network branches to *specialize* – sorting closely-related features together across distinct branches.

https://t.co/lcVBeShwow

[Distill @distillpub]

Multimodal Neurons in Artificial Neural Networks -- A new Distill article by @gabeeegoooh, @nickcammarata, @csvoss, @shancarter, @mpetrov, @ludwigschubert, @AlecRad, and @ch402.

https://t.co/IsxrQ5UCTC

[Distill @distillpub]

Self-Organising Textures — A new Distill article by @eyvindn, @zzznah, @RandazzoEttore, and @drmichaellevin. This is the third article in the Differentiable Self-organizing Systems thread.

https://t.co/dTc2406cvt

[Distill @distillpub]

Visualizing Weights — A new Distill article by @csvoss, @nickcammarata, @gabeeegoooh, @mpetrov, @ludwigschubert, Ben Egan, Swee Kiat Lim, and @ch402. This is the seventh article in the circuits thread.

https://t.co/76nzQ327qv

[Distill @distillpub]

Curve Circuits — A new Distill article by
@nickcammarata, @gabeeegoooh, @shancarter, @csvoss, @ludwigschubert, and @ch402. This is the sixth article in the circuits thread.

https://t.co/MhDgvNVRW5

[Distill @distillpub]

High-Low Frequency Detectors — A new Distill article by
@ludwigschubert, @csvoss, and @ch402. This is the fifth article in the circuits thread.
https://t.co/b8Bfl0heOo

[Distill @distillpub]

Naturally Occurring Equivariance in Neural Networks -- A new Distill article by @ch402, @nickcammarata, @csvoss, @ludwigschubert, and @gabeeegoooh. This is the fourth article in the circuits thread.

https://t.co/eeBuNI2D8g

[Distill @distillpub]

Understanding RL Vision -- A new Distill article by @JacobHHilton, @nickcammarata, @shancarter, @gabeeegoooh, and @ch402.

https://t.co/Ymcf2avEGg

[Distill @distillpub]

Communicating with Interactive Articles -- A new Distill article by @fredhohman, @mathisonian, @jeffrey_heer, and @polochau.
https://t.co/1od6Fgc3H8

[Distill @distillpub]

Interested in this line of work? Check out the #selforg channel on the Distill slack (https://t.co/BJYuSrE89f) for discussion and ongoing experiments!

[Distill @distillpub]

Self-classifying MNIST Digits -- A new Distill article by @zzznah, @RandazzoEttore, @eyvindn, @drmichaellevin and @samgreydanus. This is the second article in the differentiable self-organizing systems thread.

https://t.co/5JAQ9drKR8