LGNov 13, 2014

Greedy metrics in orthogonal greedy learning

arXiv:1411.3553v13 citations
Originality Synthesis-oriented
AI Analysis

This work addresses model selection challenges in greedy learning algorithms, offering an incremental improvement for researchers in machine learning optimization.

The paper tackles the problem of model selection in orthogonal greedy learning by introducing a new greedy metric called δ-greedy thresholds, which avoids overfitting/underfitting and provides an adaptive termination rule while maintaining learning performance.

Orthogonal greedy learning (OGL) is a stepwise learning scheme that adds a new atom from a dictionary via the steepest gradient descent and build the estimator via orthogonal projecting the target function to the space spanned by the selected atoms in each greedy step. Here, "greed" means choosing a new atom according to the steepest gradient descent principle. OGL then avoids the overfitting/underfitting by selecting an appropriate iteration number. In this paper, we point out that the overfitting/underfitting can also be avoided via redefining "greed" in OGL. To this end, we introduce a new greedy metric, called $δ$-greedy thresholds, to refine "greed" and theoretically verifies its feasibility. Furthermore, we reveals that such a greedy metric can bring an adaptive termination rule on the premise of maintaining the prominent learning performance of OGL. Our results show that the steepest gradient descent is not the unique greedy metric of OGL and some other more suitable metric may lessen the hassle of model-selection of OGL.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes