Roman Belousov

Probabilities can sometimes be counterintuitive. Is an event with probability ~ 5 × 10^-20 synonymous to impossible—never ever? Just for laughs, here is a method from Russian video blogger Lex Kravetski for doing the “impossible” at home 😁...

Post

Modeling complex systems appears to me in striking contrast with the image of physics that emerges from famous stories and anecdotes about scientific struggle: poor Galileo pacing up and down the Tower of Pisa, Newton’s revelation “beaten” into his head by an apple, desperate or uncanny hypotheses introduced by Planck, Bohr, and Einstein just to make progress. These examples give the impression, that the hardest part of the problem is to find even a single model describing the experimental observations. Today, when it comes to complex systems, suddenly there is a multitude of mathematical models for the same problem... Does it matter which one we choose?

Post

Eager to read about recent advances in symbolic machine learning unlocked by SymTorch, I nevertheless stumbled over the authors’ premise: “In physics, interpretability means using concise equations that reliably explain phenomena.” The quote reminded me of a joke: any series expansion can be called a special function, even if it is special only to your particular physical problem. If a concise equation describes the data well, does it necessarily have an obvious physical interpretation—or any at all?

Post