Learning Maximum Entropy Models from finite size datasets: a fast Data-Driven algorithm allows sampling from the posterior distribution

Abstract

Maximum entropy models provide the least constrained probability distributions that reproduce statistical properties of experimental datasets. In this work we characterize the learning dynamics that maximizes the log-likelihood in the case of large but finite datasets. We first show how the steepest descent dynamics is not optimal as it is slowed down by the inhomogeneous curvature of the model parameters space. We then provide a way for rectifying this space which relies only on dataset properties and does not require large computational efforts. We conclude by solving the long-time limit of the parameters dynamics including the randomness generated by the systematic use of Gibbs sampling. In this stochastic framework, rather than converging to a fixed point, the dynamics reaches a stationary distribution, which for the rectified dynamics reproduces the posterior distribution of the parameters. We sum up all these insights in a "rectified" Data-Driven algorithm that is fast and by sampling from the parameters posterior avoids both under- and over-fitting along all the directions of the parameters space. Through the learning of pairwise Ising models from the recording of a large population of retina neurons, we show how our algorithm outperforms the steepest descent method.

References in corpus (4)

• Proceedings of the 29th International Conference on Machine Learning (ICML-12)
• Identification of direct residue contacts in protein-protein interaction by message passing
• Dynamical criticality in the collective activity of a population of retinal neurons
• Faster solutions of the inverse pairwise Ising problem

Cited by in corpus (8)

• Inverse statistical problems: from the inverse Ising problem to data science
• Maximum entropy models reveal the excitatory and inhibitory correlation structures in cortical neuronal activity
• Separating intrinsic interactions from extrinsic correlations in a network of sensory neurons
• Modeling the correlated activity of neural populations: A review
• Resummed mean-field inference for strongly coupled data
• Pairwise Ising model analysis of human cortical neuron recordings
• The statistical mechanics of Twitter
• Linear response for spiking neuronal networks with unbounded memory