Investing tid bits..

Size of your
REWARD and TALENT

Stock market is the only place where the size of the REWARD is far higher than the TALENT

I am an above average ordinary individual. But the markets have rewarded me hugely.

All it needs - Passion, Learing & temperament

Thiruvairanikulam Mahadeva Temple
Ernakulam Kerala🚩
#temples

Temple is known as Sabarimala of Women

Thiruvairanikulam Mahadeva Temple is situated on the banks of the Periyar River. Siva Temple has a separate sannidhi for Parvathi Devi which will be open only for 12 days a year

The Nadathurappu Mahothsavam happens every year on Malayalam month of Dhanu. On opening day, a grand procession starting from Akavoor Mana carries the ornaments of Devi to the temple. Swayamvara Archana is the main pooja here to clear the obstacle in marriages.
Legend is that 👇

Vigraham existed in another place called Iranikulam. A Namboodiri of Sreemoolanagaram used to visit the Iranikulam temple regularly. Akavoor Chathan, a boatman will take him to the Iranikulam temple and back. Because of Bhakthi of Namboodiri, Shiva traveled with him to his place.

Akavoor Chathan identified Siva chaitanyam and their family built a temple here. Parvati devi herself will cook Nevidyam here for Shiva, which was witnessed by a priest. Devi was angry at this and decided to leave the place. But when the priest offered his sincere apology👇

Devi blessed him and told that she will be present here 12 days in a year. Thus the period which begins from Thiruvathirai Day in Dhanu is known as the Nadathurappu festival.
Temple Nada opens today night.
Picture credit: Google.
🙏🕉🙏

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.

There's been a lot of talk about the missionary killed by the natives of North Sentinel Island. They're probably so aggressive because of this weirdo, Maurice Vidal Portman. So here's a big thread about this creep and some facts from my decade-long obsession with the island.

The Sentinelese are often described as “uncontacted,” but this not strictly true. They had a very significant contact in 1880 with Commander Portman.

Portman, the black sheep third son of some minor noble, was assigned by the English Royal Navy to administer and pacify the Andaman Islands, a job he pursued from 1880-1900 with the full measure of his own perversity.

Portman was erotically obsessed with the Andamanese, and he indulged his passion for photography by kidnapping members of various tribes and posing them in mock-Greek homoerotic compositions.

During his 20 years in a sexualized heart of darkness, Portman measured and cataloged every inch of his prisoner’s bodies, with an obsessive focus on genitals.

Remember: Just because an idea is old doesn't mean it's good.

Knowing whether old ideas are more likely to be good requires understanding whether conditions have changed in important ways.

Horse archery was an amazing method of warfare for over a thousand years...then people invented guns, and suddenly this great idea that had stood the test of time became obsolete.

Then again, when underlying conditions don't change much, tried-and-true approaches are probably better.

"Learning the lessons of history" usually just means assuming ergodicity and stationarity in an informal time-series model...

(end)