[thread] on "sovereignty"

"Some fears have been expressed that if Britain joins the Common Market she will cease to be able to formulate her own foreign policy and will lose her separate identity. "

THREAD: 15 of the most useful razors and rules I've found.

Rules of thumb that simplify decisions.

Bezos' Razors:

• If unsure what action to take, let your 80-year-old self make it.

• If unsure who to work with, pick the person that has the best chances of breaking you out of a 3rd world prison.

Skinner's Law:

• If procrastinating on an item, you only have 2 options:

1. Make the pain of not doing it greater than the pain of doing it.

2. Make the pleasure of doing it greater than the pleasure of not doing it.

Luck Razor:

• If stuck with 2 equal options, pick the one that feels like it will produce the most luck later down the line.

I used this razor to go for drinks with a stranger rather than watch Netflix.

In hindsight, it was the highest ROI decision I've ever made.

Bragging Razor:

• If someone brags about their success or happiness, assume it’s half what they claim.

• If someone downplays their success or happiness, assume it’s double what they claim.

The map is not the terrain.

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.

Kolaru Pathigam-Thevaram hymn composed by ThiruGnana Sambandar- #Thread

In Tamil Kol means planets,Aru means cut- set of songs in Kolaru Pathigam addresses all Navagrahas& last 2 lines of every song with end a plea to Bhagwan Shiva to reduce/erase the ill effects of Navagrahas

கோளறு பதிகம் - This pathigam was sung by Thirugnana Sambandar at Thirumaraikadu Kshetram -Vedaranyam,Tamilnadu

Thirugnana Sambandar & Appar(Thirunavukkarasar) travelled together to various Shiva Alayam to spread shivam & bhakti among people.They both reached Thirumaraikadu.

Thirumaraikadu(Vedaranyam) -
Appar & Sambandar were asked to enter the temple through side entrance.The main door was locked by Vedas many years ago.Appar&Sambandar sang devotional hymns praising Shiva, after which the gates opened & closed.After which main door opened for all.

After darshan of Bhagwan Shiva at Thirumaraikkadu,Sambandar received a message to travel to Madurai.
During that time, Samanam was on the rise in the Pandya Kingdom.The Pandya King converted to Samanam.The Queen Mangaiyarkkarasi was a staunch Shiva devotee and was upset.

Kularchirai Nayanar ,Chief Minister of Pandya kingdom was in agony by the turn of events.Queen requested Kulachirai Nayanar to escort Thiru gnana Sambandar to Madurai.Kularchirai Nayanar met Thiru gnana Sambandar at Thirumaraikkadu & requested him to visit Madurai.