This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| Edwards–Anderson model | |
|---|---|
| Name | Edwards–Anderson model |
| Field | Statistical mechanics |
| Introduced | 1975 |
| Authors | Sam Edwards, Philippe W. Anderson |
| Tags | Spin glass, disordered systems |
Edwards–Anderson model The Edwards–Anderson model is a paradigmatic lattice model of spin glass behavior introduced by Sam Edwards and Philippe W. Anderson in 1975. It provides a minimal microscopic framework to study quenched disorder, frustration, and complex energy landscapes relevant to materials studied by John B. Goodenough, Nevill Mott, and experimentalists at institutions such as Bell Labs and Cambridge University. The model connects theoretical developments in Ludwig Boltzmann-inspired statistical mechanics with later conceptual advances by Pierre H. D. de Gennes, Sir Nevill F. Mott, and mathematical work influenced by Oded Schramm.
The Edwards–Anderson model was proposed in the context of anomalous magnetic alloys investigated by Philip W. Anderson and experimental studies at Bell Telephone Laboratories and Argonne National Laboratory. It formalizes interactions on regular lattices originally considered in research by Michael E. Fisher and Kenneth G. Wilson and set the stage for theoretical cross-pollination with mean-field approaches developed by Marc Mézard, Giorgio Parisi, and Miguel Ángel Virasoro. The model has been instrumental in linking numerical work from groups at Los Alamos National Laboratory and IBM Research to rigorous results pursued by mathematicians at Princeton University and École Normale Supérieure.
The Edwards–Anderson model is defined on a discrete lattice such as those used by Lars Onsager and Felix Bloch in solid-state contexts, typically a cubic or square lattice studied by L. D. Landau. On each lattice site resides an Ising spin variable σ_i ∈ {±1} as in the original model of Ernst Ising, with nearest-neighbor interactions J_ij drawn from a quenched random distribution inspired by experimental disorder characterized in work by Neal K. Ashcroft and N. David Mermin. The Hamiltonian is the classical energy function introduced by Sam Edwards and Philippe W. Anderson, paralleling techniques from John C. Slater and Lev Landau. Boundary conditions (periodic, free) are chosen as in numerical studies from Kenneth Wilson’s renormalization programs.
The model exhibits a rich phase structure echoing observations by Clifford Shull and Neils Bohr-era experiments: paramagnetic high-temperature regimes and low-temperature glassy phases with broken ergodicity studied by Giorgio Parisi, David Sherrington, and Marc Mézard. Concepts such as replica symmetry breaking developed by Giorgio Parisi and stability analyses akin to those by Ilya Prigogine and Lev Gor'kov are central to understanding the low-temperature phase. Critical phenomena described by Kenneth G. Wilson’s renormalization group appear near putative transitions, and scaling relations tested by Michael Fisher and John Cardy inform finite-size behaviors. Ultrametric and hierarchical organization conjectured by Giorgio Parisi links to mathematical structures explored by researchers at Institute for Advanced Study and University of Cambridge.
Analytical approaches include the replica method advanced by Giorgio Parisi and saddle-point techniques developed in the tradition of Richard Feynman and Julian Schwinger, while rigorous mathematical work follows paths charted by Elliott H. Lieb and Barry Simon. Mean-field analogues such as the Sherrington–Kirkpatrick model studied by David Sherrington and S. Kirkpatrick provide solvable benchmarks. Renormalization group methods from Kenneth G. Wilson, cavity approaches influenced by Marc Mézard, and variational constructions inspired by Vladimir A. Novikov and Hans Bethe are commonly employed. Probabilistic tools echo developments by Andrei Kolmogorov and Paul Erdős, and optimization perspectives relate to algorithmic work at Bell Labs and MIT.
Large-scale Monte Carlo simulations performed by groups at Los Alamos National Laboratory, ETH Zurich, and Aalto University employ algorithms such as simulated tempering from David J. Earl’s lineage and parallel tempering developed in the computational chemistry community around Martin Karplus. Finite-size scaling analyses follow methodologies established by Michael E. Fisher and Kenneth Wilson; results probe existence of a finite-temperature spin-glass transition in three dimensions debated in literature by researchers at Princeton University, University of Chicago, and Oxford University. Graphics processing unit (GPU) accelerated studies inspired by initiatives at NVIDIA and high-performance clusters at Lawrence Livermore National Laboratory have enhanced sampling, while exact ground-state computations utilize combinatorial optimization techniques from Jack Edmonds and Michael R. Garey.
Experimental realizations include dilute magnetic alloys first characterized by J. H. Van Vleck and later experiments at Bell Labs and Argonne National Laboratory, as well as artificial spin-ice systems crafted in laboratories at University of California, Santa Barbara and University of Tokyo. Applications extend to neural network models pioneered by John Hopfield, error-correcting codes influenced by Claude Shannon, and computational complexity questions discussed by Stephen Cook and Leonid Levin. Cross-disciplinary connections reach into protein folding experiments by groups associated with Stanford University and optimization routines used in industry settings at IBM Research.
Extensions include vector-spin generalizations studied by Pierre-Gilles de Gennes, p-spin glass models analyzed by Giulio Parisi’s collaborators, and diluted lattice models inspired by percolation research from Geoffrey Grimmett. Related solvable frameworks include the Sherrington–Kirkpatrick model from David Sherrington and Scott Kirkpatrick, random-field Ising models investigated by Yoseph Imry and Subir Sachdev, and hierarchical spin glasses connected to work by Benoît Mandelbrot and René Thom. Interdisciplinary offshoots tie to combinatorial optimization theory advanced by Richard Karp and to machine-learning perspectives explored at Google DeepMind and OpenAI.