Statistical physics of pairwise probability models
arXiv:0905.1410 · doi:10.3389/neuro.10.022.2009
Abstract
Statistical models for describing the probability distribution over the states of biological systems are commonly used for dimensional reduction. Among these models, pairwise models are very attractive in part because they can be fit using a reasonable amount of data: knowledge of the means and correlations between pairs of elements in the system is sufficient. Not surprisingly, then, using pairwise models for studying neural data has been the focus of many studies in recent years. In this paper, we describe how tools from statistical physics can be employed for studying and using pairwise models. We build on our previous work on the subject and study the relation between different methods for fitting these models and evaluating their quality. In particular, using data from simulated cortical networks we study how the quality of various approximate methods for inferring the parameters in a pairwise model depends on the time bin chosen for binning the data. We also study the effect of the size of the time bin on the model quality itself, again using simulated data. We show that using finer time bins increases the quality of the pairwise model. We offer new ways of deriving the expressions reported in our previous work for assessing the quality of pairwise models.
25 pages, 3 figures
References in corpus (4)
Cited by in corpus (66)
- Improved contact prediction in proteins: Using pseudolikelihoods to infer Potts models
- Searching for collective behavior in a network of real neurons
- Inverse statistical problems: from the inverse Ising problem to data science
- Inverse Ising inference using all the data
- Stimulus-dependent maximum entropy models of neural population codes
- Adaptive cluster expansion for the inverse Ising problem: convergence, algorithm and tests
- The Bethe approximation for solving the inverse Ising problem: a comparison with other inference methods
- Spin glass models for a network of real neurons
- On the criticality of inferred models
- Pseudolikelihood Decimation Algorithm Improving the Inference of the Interaction Network in a General Class of Ising Models
- Mean-field theory for the inverse Ising problem at low temperatures
- Searching for simplicity: Approaches to the analysis of neurons and behavior
- Maximum likelihood reconstruction for Ising models with asynchronous updates
- Learning and inference in a nonequilibrium Ising model with hidden nodes
- Network inference using asynchronously updated kinetic Ising Model
- Bethe-Peierls approximation and the inverse Ising model
- A unifying framework for mean-field theories of asymmetric kinetic Ising systems
- Approximate Inference for Time-varying Interactions and Macroscopic Dynamics of Neural Populations
- A message-passing scheme for non-equilibrium stationary states
- Intrinsic limitations of inverse inference in the pairwise Ising spin glass
- Quantifying Relevance in Learning and Inference
- Correlated fluctuations in strongly-coupled binary networks beyond equilibrium
- Inferring Synaptic Structure in presence of Neural Interaction Time Scales
- Inference and learning in sparse systems with multiple states
- Energy landscape analysis based on the Ising model: Tutorial review
- Dynamics and Performance of Susceptibility Propagation on Synthetic Data
- Ensemble Inhibition and Excitation in the Human Cortex: an Ising Model Analysis with Uncertainties
- Correlation-Compressed Direct Coupling Analysis
- Pairwise maximum-entropy models and their Glauber dynamics: bimodality, bistability, non-ergodicity problems, and their elimination via inhibition
- Generalized mean field approximation for parallel dynamics of the Ising model
- Reconstructing the Hopfield network as an inverse Ising problem
- A statistical physics approach to learning curves for the Inverse Ising problem
- Role of correlations in population coding
- Regularization and decimation pseudolikelihood approaches to statistical inference in -spin models
- Expansion of the effective action around non-Gaussian theories
- Inferring couplings in networks across order-disorder phase transitions
- Entropy-based parametric estimation of spike train statistics
- Translating ceRNA susceptibilities into correlation functions
- Variational approximations for stochastic dynamics on graphs
- Improved Pseudolikelihood Regularization and Decimation methods on Non-linearly Interacting Systems with Continuous Variables
- Inferring epistasis from genomic data with comparable mutation and outcrossing rate
- Optimized brute-force algorithms for the bifurcation analysis of a spin-glass-like neural network model
- Free Energy Evaluation Using Marginalized Annealed Importance Sampling
- Biases in Inverse Ising Estimates of Near-Critical Behaviour
- Granger causality and the inverse Ising problem
- Beyond inverse Ising model: structure of the analytical solution for a class of inverse problems
- Message passing algorithms for the Hopfield network reconstruction: threshold behavior and limitation
- Estimating Information-Theoretic Quantities
- Inference for interacting linear waves in ordered and random media
- Theory of population coupling and applications to describe high order correlations in large populations of interacting neurons
- A Density Consistency approach to the inverse Ising problem
- Spatial Monte Carlo Integration with Annealed Importance Sampling
- 3D Protein Structure Predicted from Sequence
- Far-from-equilibrium complex landscapes
- Inferring Higher-Order Couplings with Neural Networks
- Pairwise interactions origin of entropy functions
- Constraint-based inverse modeling of metabolic networks: a proof of concept
- Active image restoration
- Composite Spatial Monte Carlo Integration Based on Generalized Least Squares
- Empirical Bayes Method for Boltzmann Machines
- Momentum-dependence in the infinitesimal Wilsonian renormalization group
- Worldwide bilateral geopolitical interactions network inferred from national disciplinary profiles
- The Ising decoder: reading out the activity of large neural ensembles
- On the role of non-linear latent features in bipartite generative neural networks
- Solving Non-parametric Inverse Problem in Continuous Markov Random Field using Loopy Belief Propagation
- Model risk on credit risk