Pinned Tweet
ในคติพระเวท พระวรุณเป็นเทพแห่งฟากฟ้า ผู้ควบคุมจักรวาลและกฎแห่งธรรมจักร ทรงเป็นผู้รักษาสัจจะ เกลียดชังคำเท็จ ผู้ใดล่วงสัจจะจะถูกลงโทษด้วย “วรุณบาศ” หรือบ่วงแห่งพระวรุณ ซึ่งอาจก่อให้เกิดความเจ็บไข้ได้ป่วยหรือเหตุเคราะห์กรรมต่าง ๆ
แพงง่า
The classic “American Dream” — marriage, two kids, homes, healthcare, cars and education — is now an estimated $3,455,305 over the course of a lifetime, according to a new study from Investopedia.
finance.yahoo.com/news/ameri…
AlphaFold topped the CASP shared tasks and won the Nobel prize, with evaluated to the 90% line, comparable to labs, despite did not pass other criterians such as explanations.
This year’s chemistry laureates Demis Hassabis and John Jumper have developed an AI model, AlphaFold2, to solve a 50-year-old problem: predicting proteins’ complex structures.
Check out two examples of protein structures determined using AlphaFold2. First up, a bacterial enzyme that causes antibiotic resistance. The structure is important for discovering ways of preventing antibiotic resistance.
Animation: ©Terezia Kovalova/The Royal Swedish Academy of Sciences
#NobelPrize
Peratham Wiriyathammabhum retweeted
Her argument assumes the Nobel was for brute-forcing protein structures with Google’s money and compute. That’s not accurate.
The prize was for AlphaFold2’s method - an AI architecture that solves a 50-year problem pure brute force can never crack, due to the astronomical number of possible folds.
Scaling and releasing the database was ambitious, but the recognition went to the scientific breakthrough, not the resources it produced.
This is exactly why @demishassabis shouldn’t have won the Nobel. No ingenuity or thought process. Just brute force from unlimited money, unlimited compute and unlimited staff (from Google).
Peratham Wiriyathammabhum retweeted
Replying to @GaryMarcus
LLMs/AIs assisted undergrad GPAXs for years.
For over-exciting students in research, esp. hype fields like AI or Epidemiology or Physics, as growing older, the interest in these events lessen -.-
This is something.






















