# Efficient structure learning with automatic sparsity selection for causal graph processes

## TL;DR

This paper introduces a scalable, automatic sparsity selection algorithm for learning sparse causal graphs from time series data, improving efficiency and accuracy over existing methods.

## Contribution

The authors develop a cyclical coordinate descent algorithm with novel non-parametric error metrics for automatic LASSO coefficient selection in causal graph learning.

## Key findings

- Achieves state-of-the-art performance on simulated data.
- Effectively scales to dense and sparse graphs.
- Successfully applied to real stock market data.

## Abstract

We propose a novel algorithm for efficiently computing a sparse directed adjacency matrix from a group of time series following a causal graph process. Our solution is scalable for both dense and sparse graphs and automatically selects the LASSO coefficient to obtain an appropriate number of edges in the adjacency matrix. Current state-of-the-art approaches rely on sparse-matrix-computation libraries to scale, and either avoid automatic selection of the LASSO penalty coefficient or rely on the prediction mean squared error, which is not directly related to the correct number of edges. Instead, we propose a cyclical coordinate descent algorithm that employs two new non-parametric error metrics to automatically select the LASSO coefficient. We demonstrate state-of-the-art performance of our algorithm on simulated stochastic block models and a real dataset of stocks from the S\&P$500$.