The Copycat Perceptron: Smashing Barriers Through Collective Learning

Abstract

We characterize the equilibrium properties of a model of $y$ coupled binary perceptrons in the teacher-student scenario, subject to a suitable cost function, with an explicit ferromagnetic coupling proportional to the Hamming distance between the students' weights. In contrast to recent works, we analyze a more general setting in which thermal noise is present that affects each student's generalization performance. In the nonzero temperature regime, we find that the coupling of replicas leads to a bend of the phase diagram towards smaller values of $\alpha$: This suggests that the free entropy landscape gets smoother around the solution with perfect generalization (i.e., the teacher) at a fixed fraction of examples, allowing standard thermal updating algorithms such as Simulated Annealing to easily reach the teacher solution and avoid getting trapped in metastable states as it happens in the unreplicated case, even in the computationally \textit{easy} regime of the inference phase diagram. These results provide additional analytic and numerical evidence for the recently conjectured Bayes-optimal property of Replicated Simulated Annealing (RSA) for a sufficient number of replicas. From a learning perspective, these results also suggest that multiple students working together (in this case reviewing the same data) are able to learn the same rule both significantly faster and with fewer examples, a property that could be exploited in the context of cooperative and federated learning.

2 figures in the main, 5 figures in the appendix

References in corpus (20)

• The structural glass transition and the entropy of the metastable states
• Statistical physics of inference: Thresholds and algorithms
• Phase Transitions in the Coloring of Random Graphs
• Recipes for metastable states in Spin Glasses
• Unreasonable Effectiveness of Learning Neural Networks: From Accessible States and Robust Ensembles to Basic Algorithmic Schemes
• Hiding Quiet Solutions in Random Constraint Satisfaction Problems
• The simplest model of jamming
• Statistical Mechanics of Support Vector Networks
• Universality of the SAT-UNSAT (jamming) threshold in non-convex continuous constraint satisfaction problems
• Subdominant Dense Clusters Allow for Simple Learning and High Computational Performance in Neural Networks with Discrete Synapses
• Hiding solutions in random satisfiability problems: A statistical mechanics approach
• Weight Space Structure and Internal Representations: a Direct Approach to Learning and Generalization in Multilayer Neural Network
• Origin of the computational hardness for learning with binary synapses
• Efficiency of quantum versus classical annealing in non-convex learning problems
• Local entropy as a measure for sampling solutions in Constraint Satisfaction Problems
• Entropy landscape of solutions in the binary perceptron problem
• Marvels and Pitfalls of the Langevin Algorithm in Noisy High-dimensional Inference
• Limits and performances of algorithms based on simulated annealing in solving sparse hard inference problems
• Learning curves for the multi-class teacher-student perceptron
• Combined local search strategy for learning in networks of binary synapses

Cited by in corpus (1)

• Interacting Copies of Random Constraint Satisfaction Problems