I wanted to test this claim with SAT problems. Why SAT? Because solving SAT problems require applying very few rules consistently. The principle stays the same even if you have millions of variables or just a couple. So if you know how to reason properly any SAT instances is solvable given enough time. Also, it's easy to generate completely random SAT problems that make it less likely for LLM to solve the problem based on pure pattern recognition. Therefore, I think it is a good problem type to test whether LLMs can generalize basic rules beyond their training data.
Dani Barnett said she had felt she did not have anyone to talk to about menopause
。同城约会对此有专业解读
But that’s unironically a good idea so I decided to try and do it anyways. With the use of agents, I am now developing rustlearn (extreme placeholder name), a Rust crate that implements not only the fast implementations of the standard machine learning algorithms such as logistic regression and k-means clustering, but also includes the fast implementations of the algorithms above: the same three step pipeline I describe above still works even with the more simple algorithms to beat scikit-learn’s implementations. This crate can therefore receive Python bindings and even expand to the Web/JavaScript and beyond. This also gives me the oppertunity to add quality-of-life features to resolve grievances I’ve had to work around as a data scientist, such as model serialization and native integration with pandas/polars DataFrames. I hope this use case is considered to be more practical and complex than making a ball physics terminal app.
Knighthead Capital Management in early discussions
• （本文仅为作者个人观点，不代表本报立场）