I’m John Curcio, and this is my blog. I go by XTXinverseXTY on most platforms. I like weird little math problems and real data.
I’m always writing an essay or hacking on some bizarre side project, and every now and then I share what I’m working on.
Laya's Prior Art Claim is Absurd
Brief context: Jev came out recently, offering API access to a closed model that answers typed decision questions (developer-specified schemas) in one pass. SalesRLAgent (arXiv, HF repo) is an earlier model by Nandakishor Mukkunnoth that predicts sales-conversion probability from sales conversations. Laya is an open-weights, Jev-compatible alternative, released 3 days after Jev. In its launch post, Mukkunnoth claims SalesRLAgent was prior art for Jev. His claim has since circulated broadly, and many seem confused over precedence. I dug into SalesRLAgent over the weekend. I found serious errors and nothing uniquely in common with Jev. ...
KL Divergence and the Kelly Criterion
Portfolio Splitting and the Kelly Criterion Suppose the Knicks play the Spurs tomorrow. A bookmaker believes $P(\text{Knicks win}) = p$. He sells two fractional contracts, each fairly priced under that belief: A $\$1$ Knicks contract that pays $\frac{1}{p}$ if the Knicks win. A $\$1$ Spurs contract that pays $\frac{1}{1-p}$ if the Spurs win. (Note: Zero vig $\implies$ fair value = $1) A gambler believes $P(\text{Knicks win}) = q > p$. He has just $1 but wants to grow it aggressively, so he spends: ...
The Efficient Market Hypothesis is a Philosophical Razor
I’ve read some inane discussions about the validity of the efficient market hypothesis (EMH). I’ve even participated in some. Proponents warn that cocky retail traders are naive, and detractors point to historical examples of inefficiencies. I claim this is a category mistake by both parties. The EMH is more like a philosophical razor: you should accept it for a given asset, unless you have some information to the contrary. We might as well call it the efficient market null hypothesis. ...
Statistical Intuition for Elo Scores
Elo Scoring Two players have Elo scores $A$ and $B$. With no info other than that, it’s generally accepted that $P(A\ beats\ B) = 1 / (1 + 10^{(B - A) / 400})$ If A indeed does beat B, here are the new Elo scores: $A' = A + K \cdot (1 - P(A\ beats\ B))$ $B' = B - K \cdot (1 - P(A\ beats\ B))$ Where $K$ is a hyperparameter, usually something between 10 and 40. ...
Inaugural post
Hello, world! It’s me, blog!