Pinned Tweet
Training a neural network to compress elementary functions into only two numbers (x,y) on an Archimedean spiral.
From those two numbers, the original function can be recovered. Each family of functions is eventually beautifully clustered in the embedded space of the spiral.
Christos Tsanikidis retweeted
It's nearly 3 here, my favourite part of the night shift… locked in... 💪🚀
Becoming rich from working at a startup still requires beating the market while taking more individual risk than investors.
Know a guy who made $15M off the Figma IPO.
Worked there 5 years, got in at $300M valuation and initial comp was 75k/yr of equity
Your reminder it doesn't take founding something to get rich.
3
577
Here’s a trading game where you are betting on the fair value of barrier contracts on 2D brownian motions hitting bubbles.
Go long if the market % is too low, short otherwise.
1
2
387
How long before frontier AI labs start taking **some** business away from Hedge Funds and Prop shops?
Edge tied to insanely hard-to-get data will likely remain for several years.
Research is like a diffusion out of a labyrinth.
Each researcher is a particle that faces obstacles in their effort to expand the frontier, yet they may bias their random walk around the challenges.
3
346
Simulated Annealing on the Maximum Independent set Problem is very cool.
The random solution at any time follows a random walk in the space of valid graphs. At the same time, its objective value itself is very Brownian-motion like.
2
542
Sampling the US map with Markov Chains results in a pleasing animation when the step size is large enough.
1
6
610
Markov Chain based sampling: sampling from various distributions using MCMC and Metropolis-Hastings.
more: tsan.me/post/mcmc/
Luck often narrows edges. Impact of your skill versus impact of randomness (the SNR ratio), should be a determining factor in taking bets.
Suppose your life depends on beating either the chess champion (@MagnusCarlsen) or the poker champion (choose one). What should you choose?
Replying to @thegautamkamath
Suppose N papers with novelty nᵢ~𝒩(0,1) are submitted and N/4 are accepted. Assume the novelty of your paper is n, and you judge it to be 𝒩(n,0.5) – and reviewers judge it to be 𝒩(n,0.8). Even if you think you are top 10% your chance of acceptance can be as low as ≈64%
2
3
What does Strong Duality really tell us: All these complicated constraints can be *summarized* by a single plane! It’s remarkable that such a reduction is possible (when it is / under mild conditions in convex programs).
5
In chess, they say when you find a good move, look for a better one. @nntaleb mentions a related “contamination principle” in the value of financial contracts.
This “contamination” often holds true in research too. Eg when you have a good algorithm, look for a better one.
2
The (a) *Gaussian distribution* and (b) *Convexity (of sets)* are in some sense, similarly universal. How? (a) Summing non-gaussian r.v. often gives you approximately a gaussian rv. (b) Summing non-convex sets gives you approximately a convex set*. figure: arxiv.org/pdf/1712.08559.pdf
1
1
exp(x) has the interpretation of an x/n percentage increase, n times (n → ∞) (continuous compounding). Here’s a continuous interpretation of the logarithm: log(x) is the continuous percentage increase from 1 to x.
1
3










