#Gold $Gold
Next Breakout will trigger above $1891
for $1960 around https://t.co/JG9dlhTj64

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.

#Thread The legend of Ekadash Mukhi Hanuman ji...

Mahadev himself narrated the story of Ekadash Mukhi Hanuman ji to Maa Parvati. Once a great monster named Kalkarmukh performed the severe penance of Brahma ji.

Pleased with the austerity of the demon, Brahma ji gave him a boon that the person who will have eleven Heads like you and who was born on the day of your birth, only he/she will be able to kill you. No one else will kill you.

After getting a boon from Brahma ji, Kalkarmukh created a ruckus in all three lok. He invaded Swarg and snatched all the rights of the Devtas and became himself the king of Swarg.

All Devtagan came to Shri Rama and asked him for help. Hearing the agony of the Devtagan, Shri Ram called his supreme devotee Hanuman ji and said, "Pawan Putra, Sankatmochan".

1) My Stan Lee story happened when I was in 5th grade. I was visiting NYC with my Dad and had read the address for Marvel in the comics I read constantly. I begged my Dad to take me and to his credit he did. And guess what? It was just an office.

2) We were down in the lobby and I was kind of crushed. I don’t know what I expected, The Thing and the Yancy Street Gang to be sitting around smoking cigars? Anyway, my Dad was taking a beat to figure where we were going next and a guy came up to us.

3) He was wearing a white shirt and tie and said to Dad “Is he disappointed because. The Marvel offices were just offices?” My Dad said yes and then the guy who had gray around his temples and a mustache said “hold on a second” and opened one of those office mailboxes with a key.

4) He then handed me a thick stack of EVERY SINGLE MARVEL COMIC COMING OUT THE NEXT MONTH. “Here you go. Keep reading Marvel comics” he said and then walked off. I left in a daze and about 15 minutes later it hit me “Gray around the temples, mustache... That was Stan Lee!”

5) Later when I wrote on the Ant Man movie I told Kevin Feige the story, the year, look of the guy etc and Kevin said “That’s exactly the kind of thing Stan would do and he would have been there then. That was him.” Rest In Peace Stan Lee and thank you for the comics.

Хајде да направимо мали осврт на случај Мика Алексић .

Алексић је жртва енглеске освете преко Оливере Иванчић .
Мика је одбио да снима филм о блаћењу Срба и мењању историје Срба , иза целокупног пројекта стоји дипломатски кор Британаца у Београду и Оливера Иванчић

Оливера Илинчић је иначе мајка једне од његових ученица .
Која је претила да ће се осветити .

Мика се налази у притвору због наводних оптужби глумице Милене Радуловић да ју је наводно силовао човек од 70 година , са три бајпаса и извађеном простатом пре пет година

Иста персона је и обезбедила финансије за филм преко Беча а филм је требао да се бави животом Десанке Максимовић .
А сетите се и ко је иницирао да се Десанка Максимовић избаци из уџбеника и школства у Србији .

И тако уместо романсиране верзије Десанке Максимовић утицај Британаца

У Србији стави на пиједестал и да се Британци у Србији позитивно афирмишу како би се на тај начин усмерила будућност али и мењао ток историје .
Зато Мика са гнушањем и поносно одбија да снима такав филм тада и почиње хајка и претње која потиче из британских дипломатских кругова

Најгоре од свега што је то Мика Алексић изговорио у присуству високих дипломатских представника , а одговор је био да се све неће на томе завршити и да ће га то скупо коштати .
Нашта им је Мика рекао да је он свој живот проживео и да могу да му раде шта хоће и силно их извређао

A thread on HN about bad code in legacy projects both makes me think how little we've learned as a discipline over the years and, honestly, how little credit we give ourselves for some pretty major

Fun going down this list and thinking: "Hmm, plausible at a well-run modern software shop", "Hmm, possible, but requires implausible tradeoffs", "Literally disallowed by languages", and "If you were to attempt doing that our test suite wouldn't let you merge."

I think we as an industry celebrate (not quite the right word) failure too much and don't celebrate success nearly enough. There is no DailyWTF for competent execution, word of which generally stays pretty local to the source while incompetence passes into legend.

Alrighty let me try to thread the needle on being the change I want to see in the world while not giving away anything that will get me in trouble:

Ruby has wonderful developer ergonomics. Typed languages are easier for machines to guarantee the correctness of. We built a type checker for Ruby (and I believe it is slated for OSS release sometime).

c.f.