Information theory

By Kevin P. Murphy. 2006

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Protection Against Reconstruction and Its Applications in Private Federated Learning (2018, neutral): “Claim A.2 (Lemma 4.7.2 in [9]) .”
Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces. (2015, methodology): “The forces on atoms are predicted by Bayesian inference using standard GP regression [16,23].”
The second laws of quantum thermodynamics (2013, unknown): “A construction of p′′ can be found in [53] for α > 1.”
Sharing Attention Weights for Fast Transformer (2019, methodology): “To verify this, we compute the Jensen-Shannon (JS) divergence to measure how the attention weight distribution of a layer is different from another [Lin, 1991].”
Neural Network Renormalization Group (2018, background): “, the unnormalized probability density πðxÞ of a physical problem, direct sampling of the physical configurations is generally difficult due to the intractable partition function Z 1⁄4 R dxπðxÞ [61].”
Compressive Sensing and Low-Rank Libraries for Classification of Bifurcation Regimes in Nonlinear Dynamical Systems (2013, background): “Compressive sensing may also play a role in similar statistical learning, library-based, and/or information theory methods [15, 7] used in fluid dynamics [8, 2], climate science [24, 7], and oceanography [1].”
Simulation-based optimal Bayesian experimental design for nonlinear systems (2011, background): “This quantity is non-negative, non-symmetric, and reflects the difference in information carried by the two distributions (in units of nats) [30, 31].”
Ozone profile and tropospheric ozone retrievals from the Global Ozone Monitoring Experiment: Algorithm description and validation (2005, background): “[14] According to the Nyquist criterion [Goldman, 1953], substantial spectral undersampling occurs in band 1 and in the spectral region of 325–342 nm (band 2) because the spectral sampling rate for GOME is 0.11 nm/pixel while the spectral resolution is 0.10–0.20 nm (Figure 2).”
Quantum state discrimination and its applications (2015, background): “Note that they were already obtained in earlier works by Holevo and Yuen [8, 7, 9].”
Unsupervised Learning and Segmentation of Complex Activities from Video (2018, methodology): “We draw ρk using slice sampling [15] from the conjugate prior distribution PGMM0 according to Eq.”
Unifying framework for relaxations of the causal assumptions in Bell's theorem. (2014, background): “Via the Pinsker inequality [30, 31] and the definition of mutual information (see eq.”