Stochastic Relaxation

2014

449 citations recorded by the source corpus.

Subfields: Statistical and Nonlinear Physics, Statistics and Probability

Citations in the corpus

• On the Anatomy of MCMC-based Maximum Likelihood Learning of Energy-Based Models (2019, background): “where Z` ∼ N(0, IN ) and ε > 0, has stationary distribution pθ [8, 22].”
• DeepWalk: online learning of social representations (2014, methodology): “Relational classification (or collective classification) methods [15, 25, 32] use links between data items as part of the”
• Divergence Triangle for Joint Training of Generator Model, Energy-Based Model, and Inferential Model (2018, methodology): “Both expectations can be approximated by MCMC, such as Gibbs sampling [11], Langevin dynamics, or Hamiltonian Monte Carlo (HMC) [34].”
• A survey on deep learning for big data (2018, methodology): “[13] used some specialized strategies in their developed learning framework [66,67], For instance, they put the parameters and some training objects into the global memory to reduce the data transfer.”
• Building a Telescope to Look Into High-Dimensional Image Spaces (2018, methodology): “Gibbs distributions defined on a pixel graph have been widely used as an effective family for modeling patterns of real images [13, 17, 18, 52, 53].”
• Scattering Representations for Recognition (2013, background): “If X[n] is a Markov random field then its probability distribution is a Gibbs distribution [GG84].”
• Bayesian reasoning and machine learning (2012, unknown): “5 Lattice Models Undirected models have a long history in different branches of science, especially statistical mechanics on lattices and more recently as models in visual processing in which the models encourage neighbouring variables to be in the same states[35, 36, 106].”
• Recommender Systems (2012, methodology): “Well developed inference methods including likehood maximization [180] and Gibbs sampling [181] can thus be employed in pLSA.”
• Multi-scale context for scene labeling via flexible segmentation graph (2016, background): “All the baselines are divided into two categories: (1) Modeling scene labeling using CRFs or MRFs, including TextonBoost (TB, [35]), relative location prior (RL, [37]), hierarchical CRF (HCRF, [42]), dynamic hybrid MRF (DHM, [23]), and full-connected CRF (FCRF, [59]); (2) Modeling scene labeling usi”
• Optical flow modeling and computation: A survey (2015, background): “Among early methods for minimizing (41), simulated annealing [97] offers proved convergence towards the global minimum based on stochastic relaxation.”
• Random tree walk toward instantaneous 3D human pose estimation (2015, methodology): “In the same spirit as Markov Chain Monte Carlo (MCMC) optimization approaches [12], we rely on the stochastic relaxation in choosing the unit direction.”