Regularization via f-Divergence: An Application to Multi-Oxide Spectroscopic Analysis

TL;DR

This paper introduces a novel f-divergence based regularization technique for CNNs to improve multi-oxide spectroscopic analysis of planetary surfaces, effectively reducing overfitting and enhancing prediction accuracy.

Contribution

The paper proposes a differentiable f-divergence regularizer for neural networks, specifically tailored for multi-target regression in planetary surface composition analysis, demonstrating improved performance over standard methods.

Findings

f-divergence regularization outperforms L1, L2, and dropout methods.

Combining f-divergence with standard regularizers yields the best results.

Method is validated on Mars-like spectral data from rover instruments.

Abstract

In this paper, we address the task of characterizing the chemical composition of planetary surfaces using convolutional neural networks (CNNs). Specifically, we seek to predict the multi-oxide weights of rock samples based on spectroscopic data collected under Martian conditions. We frame this problem as a multi-target regression task and propose a novel regularization method based on f-divergence. The f-divergence regularization is designed to constrain the distributional discrepancy between predictions and noisy targets. This regularizer serves a dual purpose: on the one hand, it mitigates overfitting by enforcing a constraint on the distributional difference between predictions and noisy targets. On the other hand, it acts as an auxiliary loss function, penalizing the neural network when the divergence between the predicted and target distributions becomes too large. To enable backpropagation during neural network training, we develop a differentiable f-divergence and incorporate it into the f-divergence regularization, making the network training feasible. We conduct experiments using spectra collected in a Mars-like environment by the remote-sensing instruments aboard the Curiosity and Perseverance rovers. Experimental results on multi-oxide weight prediction demonstrate that the proposed $f$-divergence regularization performs better than or comparable to standard regularization methods including $L_1$, $L_2$, and dropout. Notably, combining the $f$-divergence regularization with these standard regularization further enhances performance, outperforming each regularization method used independently.