In 2009, an unknown investor from Russia came out of nowhere to make one of the greatest late stage venture bets in the history of Silicon Valley. Why did FB choose an outsider to lead our Series D round? Here’s the inside story:

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.

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.

#Guruvayur_Krishna_Bhakthi  🕉
#Guruvayur
#Storytime
Beautiful Story of Guruvayurappan

Poonathanam Namboothiri was a great devotee of Guruvayurappan. One day Poonthanam was going to Guruvayur Kshetram. It was night time and eventhough Namboothiri was tired he continued to walk

Suddenly some thieves came and blocked his way. Poonthanam cried for Shri Krishna to help. Thieves snatched all valuables from him, but Namboothiri holding his Bhaghavatam book called Shri Krishna to help. That time a man came in a horse back and fought with thieves and protected

Namboothiri. Poonthanam thanked Krishna for helping him from danger and said to the man that "Shri Krishna has send you here and I don't know how to thank you".  The man smiled and told that he is Mangattachan, minister of  samoodhiris. Poonthanam once again thanked him and

gave his ring to Mangattachan as a token of thanks and continued his journey. He reached Guruvayur early morning for Nirmalya Darshan. Temple poojari came near him and told that Guruvayurappan came in his dreams and asked to hand over the ring to poonthanam.

When he opened Kshetra Nada poonthanam was surprised to see his ring on Guruvayurappan, which he gifted to Mangattachan. He understood that it was Guruvayurappan who came as Mangattachan to protect him. This is how Bhagwan treat his devotees.

Hare Krishna
🙏🕉🙏

Here is my Thread on 'Ketu (South Node) Through Nakshatras' The 27 Constellations 🙏❣.

Do check your “Past life Mastery” which KETU Represents🤞🏻.

1. Ketu in Ashwini Nakshatra:

Here native is Great at Financial Planning, Face reading, understanding other’s mind and intention, Driving Fast they don’t like but they are wayyy smooth over driving Fast. This is also great position which can make native Great Surgeon,

Neurologist, Psychologist.
These natives are Selfless and Sacrifice self for others. They gets Fame and Success but they just don’t care.

2. Ketu in Bharani Nakshatra:

They are good at Flirting, Attracting opposite sex, studing ancient subjects. They are also Great at Creativity and Artistic activities. The are lazy but they are really open minded who handles situations really well.

In Past life they have handles many things which is now helping them to make things easier in this life. They are great Leaders too.

They are really good at Hard work and Smart work both. And yes just as Ashwini they don’t much care about Fame and Success!

SRI VARAHA NARSIMHA, SIMHACHALAM, VIZAG (AP)
#bharatmandir

One of the 32 Narsimha Kshetram of Andhra. It has a unique form of Narsimha, appearing like a shivlingam

It is said that Narsimha’s fierce nature is soothed by worshipping him with Varaha, considered a peaceful deity

The shrine is believed to have been constructed by Prahlada. With time, it crumbled.
In Treta yuga, King Pururava was passing from here in his vimana. The vimana was attracted to this place by a mystic power.
When the place was unearthed, the deity was found.
@GunduHuDuGa

Pururava’s wife, Urvashi, had a dream that the deity should remain covered in sandalwood paste for the whole year except on Akshaya Tritiya.
This practice has continued till date.
The mukhya mandap here, has a pillar kappam stambham believed to possess curative powers.

Chandanotsava/ Chandan Yatra is the most important festival celebrated on Akshaya Tritiya. The sandalwood paste is removed from the moolavar deity and nijarupa darshan (original form of deity) for twelve hours of the day are done by devotees.