Learning single-index models via harmonic decomposition
By: Nirmit Joshi , Hugo Koubbi , Theodor Misiakiewicz and more
Potential Business Impact:
Finds hidden patterns in data by looking from all sides.
We study the problem of learning single-index models, where the label $y \in \mathbb{R}$ depends on the input $\boldsymbol{x} \in \mathbb{R}^d$ only through an unknown one-dimensional projection $\langle \boldsymbol{w}_*,\boldsymbol{x}\rangle$. Prior work has shown that under Gaussian inputs, the statistical and computational complexity of recovering $\boldsymbol{w}_*$ is governed by the Hermite expansion of the link function. In this paper, we propose a new perspective: we argue that "spherical harmonics" -- rather than "Hermite polynomials" -- provide the natural basis for this problem, as they capture its intrinsic "rotational symmetry". Building on this insight, we characterize the complexity of learning single-index models under arbitrary spherically symmetric input distributions. We introduce two families of estimators -- based on tensor unfolding and online SGD -- that respectively achieve either optimal sample complexity or optimal runtime, and argue that estimators achieving both may not exist in general. When specialized to Gaussian inputs, our theory not only recovers and clarifies existing results but also reveals new phenomena that had previously been overlooked.
Similar Papers
Neural Networks Learn Generic Multi-Index Models Near Information-Theoretic Limit
Machine Learning (Stat)
Teaches computers to learn hidden patterns faster.
Robustly Learning Monotone Single-Index Models
Machine Learning (CS)
Teaches computers to learn from messy, tricky data.
Survey on Algorithms for multi-index models
Machine Learning (Stat)
Finds patterns faster in big data.