A New Look at an Old Problem: A Universal Learning Approach to Linear Regression

Abstract

Linear regression is a classical paradigm in statistics. A new look at it is provided via the lens of universal learning. In applying universal learning to linear regression the hypotheses class represents the label $y\in \mathcal{R}$ as a linear combination of the feature vector $x^T\theta$ where $x\in {\mathcal{R}}^M$, within a Gaussian error. The Predictive Normalized Maximum Likelihood (pNML) solution for universal learning of individual data can be expressed analytically in this case, as well as its associated learnability measure. Interestingly, the situation where the number of parameters $M$ may even be larger than the number of training samples $N$ can be examined. As expected, in this case learnability cannot be attained in every situation; nevertheless, if the test vector resides mostly in a subspace spanned by the eigenvectors associated with the large eigenvalues of the empirical correlation matrix of the training data, linear regression can generalize despite the fact that it uses an ``over-parametrized'' model. We demonstrate the results with a simulation of fitting a polynomial to data with a possibly large polynomial degree.

References in corpus (2)

• Deep pNML: Predictive Normalized Maximum Likelihood for Deep Neural Networks
• Universal Supervised Learning for Individual Data

Cited by in corpus (15)

• Benign overfitting in ridge regression
• Memorizing without overfitting: Bias, variance, and interpolation in over-parameterized models
• Optimal Regularization Can Mitigate Double Descent
• Optimal ridge penalty for real-world high-dimensional data can be zero or negative due to the implicit ridge regularization
• Deep pNML: Predictive Normalized Maximum Likelihood for Deep Neural Networks
• Amortized Conditional Normalized Maximum Likelihood: Reliable Out of Distribution Uncertainty Estimation
• Increasing Depth Leads to U-Shaped Test Risk in Over-parameterized Convolutional Networks
• Single Layer Predictive Normalized Maximum Likelihood for Out-of-Distribution Detection
• The Geometry of Over-parameterized Regression and Adversarial Perturbations
• Utilizing Adversarial Targeted Attacks to Boost Adversarial Robustness
• Distribution Free Uncertainty for the Minimum Norm Solution of Over-parameterized Linear Regression
• Efficient Data-Dependent Learnability
• Bias-variance decomposition of overparameterized regression with random linear features
• Double Descent Optimization Pattern and Aliasing: Caveats of Noisy Labels
• Mitigating deep double descent by concatenating inputs