Ion Exchange - to be in the watchlist, once it crosses the resistance line than the journey towards 4K will start

Updated chart

Cheers https://t.co/tJZWZZp0SE

So here is a secret: Not everything I write goes anywhere. I have at least 4 books completely written and edited that never went on submission. I have books that stalled out at 20k. I have docs on my computer with opening paragraphs I don't remember writing. But! 1/

Every now and then an idea sticks with me like that corner of tortilla chip that just won't go down your throat. It happened with SPARROWHAWK, which began as 40k of a YA story, died, and was reborn as a comic, thanks to @christopher_j_r and @boomstudios. And now... 2/

It's happening with THE WILLOWS, a short story that you'll see in @UncannyMagazine next year. This Southern Gothic horror story began as an attempt to sell my next Romance to @AbZurdity after the Blud books in 2014 or so. And she didn't buy it. Guess why? 3/

Turns out, it wasn't a Romance! It was Southern Gothic Horror. Even the bones of it screamed strangeness, not sexiness. I loved the taste of the world, but I hadn't found the right story yet. Tried to write it 2 more times and failed. Until I finally understood what it was. 4/

So I took the 50 pages I'd written and cut half of them, twisting the story around from a gothy Romance inspired by the band the Civil Wars and turned it into a spooky descent into madness. And it looked back at me from the abyss as if to say, DUH. Sometimes... 5/

Important paper from Google on large batch optimization. They do impressively careful experiments measuring # iterations needed to achieve target validation error at various batch sizes. The main "surprise" is the lack of surprises. [thread]

https://t.co/7QIx5CFdfJ

The paper is a good example of lots of elements of good experimental design. They validate their metric by showing lots of variants give consistent results. They tune hyperparamters separately for each condition, check that optimum isn't at the endpoints, and measure sensitivity.

They have separate experiments where the hold fixed # iterations and # epochs, which (as they explain) measure very different things. They avoid confounds, such as batch norm's artificial dependence between batch size and regularization strength.

When the experiments are done carefully enough, the results are remarkably consistent between different datasets and architectures. Qualitatively, MNIST behaves just like ImageNet.

Importantly, they don't find any evidence for a "sharp/flat optima" effect whereby better optimization leads to worse final results. They have a good discussion of experimental artifacts/confounds in past papers where such effects were reported.

THE BATTLE OF PRATAPGARH (10 Nov, 1659)

Chhatrapati Shivaji Maharaj's impeccable military genius and matchless courage humbled the Adil Shahi Sultanate.

The beginning of what was to come later for the M0ghuls from one of the most indefatigable defenders of Dharma!

(Thread)

Shivaji’s growing influence in Maval region, made him a threat to the Adil Shahi ruler of Bijapur who sought to restrain him. The man who was given the responsibility of this mission, was their commander, Afzal Khan. Afzal was a formidable warrior known for his ruthlessness.(1)

His strategy was to bring Shivaji out into the plains, where he held an advantage, compared to the rocky Deccan terrain. Afzal attacked the holy town of Pandharpur, and later demolished the temple of Bhavani at Tuljapur to instigate Shivaji in a bid to bring him out.(2)

Shivaji was angry but remained patient. Knowing that he could not defeat Bijapuri Sultanate in an open battle, Shivaji moved to Pratapgarh fort, located on rocky heights, surrounded by thick forests of Jawali.(3)

This set the pattern for future Maratha attacks, using guerrilla warfare against larger invading armies rendering them ineffective in unfavorable terrain.(4)

I'm going to attempt to chronologically list all of the Marvel characters Stan Lee had a hand in creating. Working with his genius collaborators, Stan and his team produced an insane amount of iconic characters in a roughy ten year period. (Thread)

With his younger brother, writer Larry Lieber, and Jack "the King" Kirby:

Groot

Tales to Astonish #13 (November, 1960)

With writer Lieber, and artist Kirby:

Fin Fang Foom

Strange Tales #89 (October, 1961)

With the King, Stan gives birth to the Marvel Universe:

The Fantastic Four

Fantastic Four #1 (November, 1961)

With Jack Kirby:

Mister Fantastic

Fantastic Four #1 (November, 1961)

So here is a secret: Not everything I write goes anywhere. I have at least 4 books completely written and edited that never went on submission. I have books that stalled out at 20k. I have docs on my computer with opening paragraphs I don't remember writing. But! 1/

Every now and then an idea sticks with me like that corner of tortilla chip that just won't go down your throat. It happened with SPARROWHAWK, which began as 40k of a YA story, died, and was reborn as a comic, thanks to @christopher_j_r and @boomstudios. And now... 2/

It's happening with THE WILLOWS, a short story that you'll see in @UncannyMagazine next year. This Southern Gothic horror story began as an attempt to sell my next Romance to @AbZurdity after the Blud books in 2014 or so. And she didn't buy it. Guess why? 3/

Turns out, it wasn't a Romance! It was Southern Gothic Horror. Even the bones of it screamed strangeness, not sexiness. I loved the taste of the world, but I hadn't found the right story yet. Tried to write it 2 more times and failed. Until I finally understood what it was. 4/

So I took the 50 pages I'd written and cut half of them, twisting the story around from a gothy Romance inspired by the band the Civil Wars and turned it into a spooky descent into madness. And it looked back at me from the abyss as if to say, DUH. Sometimes... 5/