Tag
This paper explores using a multimodal LLM to generate domain-specific feature pools for time series anomaly detection, showing that a simple statistical pipeline can match the performance of advanced pretrained models on benchmark datasets.
The article discusses the limitations of using LLMs directly as classifiers and proposes treating LLM outputs as features in traditional machine learning models like logistic regression to achieve better calibration, interpretability, and performance.
An intern boosted a classification model's performance to 99% using label hashing, which turned out to be data leakage, leading to ethical discussions and a review of the team's practices.
A controlled study finds that feature engineering gains diminish for stronger tabular foundation models, while adding in-context information from related datasets still improves performance.
py-evoFE is an open-source Python library that uses genetic algorithms to automate and optimize feature engineering for tabular machine learning datasets, with scikit-learn compatibility and Polars for performance.
The paper introduces Crafter, an agent for corrective feature discovery that mines the residual of frozen black-box forecasters using compositional search and LLM-generated features, achieving up to 27% error reduction across six datasets and backbones.
A Kaggle Grandmaster is offering a free course covering the entire machine learning process, including EDA, decision trees, gradient boosting, feature engineering, time series, and competition strategies.
This arXiv preprint challenges the 'Garbage In, Garbage Out' heuristic, arguing that aggressive manual data cleaning can limit predictive performance in high-dimensional tabular data by reducing dimensionality needed to triangulate latent drivers.