The simplest model of jamming
From MaRDI portal
Abstract: We study a well known machine learning model -the perceptron- as a simple model of jamming of hard objects. We exhibit two regimes: 1) a convex optimisation regime where jamming is hypostatic and non-critical. 2) a non convex optimisation regime where jamming is isostatic and critical. We characterise the critical jamming phase through exponents describing the distributions law of forces and gaps. Surprisingly we find that these exponents coincide with the corresponding ones recently computed in high dimensional hard spheres. In addition, modifying the perceptron to a random linear programming problem, we show that isostaticity is not a sufficient condition for singular force and gap distributions. For that, fragmentation of the space of solutions (replica symmetry breaking) appears to be a crucial ingredient. We hypothesise universality for a large class of non-convex constrained satisfaction problems with continuous variables.
Recommendations
- Jamming: a simple introduction
- Jamming model for the extremal optimization heuristic
- A model of irreversible jam formation in dense traffic
- The jammed phase of the Biham-Middleton-Levine traffic model
- On the study of jamming percolation
- scientific article; zbMATH DE number 2095951
- Simple analytical models for the J-lay problem
Cited in
(31)- Landscape and training regimes in deep learning
- Algorithmic pure states for the negative spherical perceptron
- Following the evolution of glassy states under external perturbations: the full replica symmetry breaking solution
- The jamming transition in high dimension: an analytical study of the TAP equations and the effective thermodynamic potential
- Phase transitions in integer linear problems
- Jamming model for the extremal optimization heuristic
- Triple descent and the two kinds of overfitting: where and why do they appear?*
- Pattern capacity of a single quantum perceptron
- Storage capacity in symmetric binary perceptrons
- Gradient descent dynamics and the jamming transition in infinite dimensions
- Mean field theory of jamming of nonspherical particles
- Universality of jamming of nonspherical particles
- Out-of-equilibrium dynamical mean-field equations for the perceptron model
- Self-planting: digging holes in rough landscapes
- Scaling description of generalization with number of parameters in deep learning
- Surfing on minima of isostatic landscapes: avalanches and unjamming transition
- Critical properties of the SAT/UNSAT transitions in the classification problem of structured data
- Dynamical mean-field theory and aging dynamics
- A jamming transition from under- to over-parametrization affects generalization in deep learning
- A continuous constraint satisfaction problem for the rigidity transition in confluent tissues
- Quenched complexity of equilibria for asymmetric generalized Lotka–Volterra equations
- Tractability from overparametrization: the example of the negative perceptron
- Dynamical mean field theory for models of confluent tissues and beyond
- Capacity lower bound for the Ising perceptron
- On the topology of solutions to random continuous constraint satisfaction problems
- Exact full-RSB SAT/UNSAT transition in infinitely wide two-layer neural networks
- Graphical model for factorization and completion of relatively high rank tensors by sparse sampling
- A CLuP algorithm to practically achieve 0.76 SK-model ground state free energy
- Dynamical loss functions shape landscape topography and improve learning in artificial neural networks
- Overlap gap and computational thresholds in the square wave perceptron
- Fragile vs robust multiple equilibria phases in generalized Lotka-Volterra model with non-reciprocal interactions
This page was built for publication: The simplest model of jamming
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3186436)
