Category: Uncategorized

Suffering in fundamental physics?

a really fun paper, mind expandingly strange:

This essay explores the speculative possibility that fundamental physical operations — atomic movements, electron orbits, photon collisions, etc. — could collectively deserve significant moral weight. While I’m personally doubtful about this, I suggest reasons to keep an open mind on the topic. In practice I might adopt a kind of moral-pluralism approach in which I maintain some concern for animal-like beings even if numerically, simple physics-based suffering dominates. I also explore whether, if the multiverse does contain enormous amounts of suffering from fundamental physical operations, there are ways we can change how much of it occurs and what distribution of “experiences” it entails. An argument based on vacuum fluctuations during the eternal lifetime of the universe suggests that if we give fundamental physics any nonzero weight, then almost all of our expected impact may come through how intelligence might transform fundamental physics to reduce the amount of suffering it contains.

Predicting mutations

soon it may even be possible to make evolutionary forecasts. Scientists may not be able to predict what life will be like 100 million years from now, but they may be able to make short-term forecasts for the next few months or years. And if they’re making predictions about viruses or other health threats, they might be able to save some lives in the process.

a procedure to predict influenza mutations every year with 90% precision, helping to produce the right vaccines for that season.

We can analyze any mutation, even those that haven’t been identified yet. This allows researchers to predict whether a novel mutation is likely to be dangerous or harmless — in essence, performing a screening test

and another one

10 years of work at Johns Hopkins has yielded a computer program that predicts, with far more accuracy than current methods, which mutations are likely to have the largest effect on the activity of the “dimmer switches” (which alter the cell’s gene activity) in DNA — suggesting new targets for diagnosis and treatment of many diseases.

Massive imagenet progress

Accuracy improved to 43.9%, from 22.5% last year, and the error rate fell to 6.6%, from 11.7%. Since the Imagenet Challenge began in 2010, the error rate has decreased 4x.

mind-boggling progress in just 1 year. for comparison, humans have ~97% accuracy on these tasks. you can see them here