Results 71 to 80 of about 424,964 (284)
Reproducibility and Natural Exponential Families with Power Variance Functions
The Annals of Statistics, 1986 Let \(X_ 1\),..., \(X_ n\) be independent identically distributed random variables whose common distribution belongs to a family \({\mathcal F}=\{F_{\theta}\in \Theta \subset {\mathbb{R}}\}\) indexed by a parameter \(\theta\). We say that \({\mathcal F}\) is reproducible if there exists a sequence \(\{\) \(\alpha\) (n)\(\}\) such that \[ {\mathcal L ...
Bar-Lev, Shaul K., Enis, Peter
openaire +2 more sources
A Scalable Perovskite Platform With Multi‐State Photoresponsivity for In‐Sensor Saliency Detection
Advanced Materials, EarlyView.A scalable in‐sensor computing platform (32 × 32 array) with ultra‐low variability is developed by incorporating ferroelectric copolymers into halide perovskite thin films. These devices achieve 1000 programmable photoresponsivity states and high thermal reliability.
Xuechao Xing +10 more
wiley +1 more source
Exponential Family Hybrid Semi-Supervised Learning [PDF]
, 2009 We present an approach to semi-supervised learning based on an exponential family characterization. Our approach generalizes previous work on coupled priors for hybrid generative/discriminative models. Our model is more flexible and natural than previous
Agarwal, Arvind, Daume III, Hal
core +4 more sources
Minimum Average Deviance Estimation for Sufficient Dimension Reduction
, 2016 Sufficient dimension reduction reduces the dimensionality of data while preserving relevant regression information. In this article, we develop Minimum Average Deviance Estimation (MADE) methodology for sufficient dimension reduction.
Adragni, Kofi P. +2 more
core +1 more source
Advanced Science, EarlyView.We introduce AutomataGPT, a generative pretrained transformer (GPT) trained on synthetic spatiotemporal data from 2D cellular automata to learn symbolic rules. Demonstrating strong performance on both forward and inverse tasks, AutomataGPT establishes a scalable, domain‐agnostic framework for interpretable modeling, paving the way for future ...
Jaime A. Berkovich +2 more
wiley +1 more source
Horizon-Independent Optimal Prediction with Log-Loss in Exponential Families
, 2013 We study online learning under logarithmic loss with regular parametric models. Hedayati and Bartlett (2012b) showed that a Bayesian prediction strategy with Jeffreys prior and sequential normalized maximum likelihood (SNML) coincide and are optimal if ...
Bartlett, Peter +4 more
core
Objective Improvement in Information-Geometric Optimization [PDF]
, 2013 Information-Geometric Optimization (IGO) is a unified framework of stochastic algorithms for optimization problems. Given a family of probability distributions, IGO turns the original optimization problem into a new maximization problem on the parameter ...
Akimoto, Youhei, Ollivier, Yann
core +3 more sources
Estimation in spin glasses: A first step
, 2006 The Sherrington--Kirkpatrick model of spin glasses, the Hopfield model of neural networks and the Ising spin glass are all models of binary data belonging to the one-parameter exponential family with quadratic sufficient statistic.
Chatterjee, Sourav
core +4 more sources
On the Origins of Toughness in Corymbia calophylla (Marri Tree) Nuts
Advanced Science, EarlyView.We uncover the natural toughening mechanisms of the marri nut, including fiber pullout, crack deflection, and a viscoelastic matrix, which enable exceptional energy absorption and ductility comparable to Teflon, with an elastic modulus similar to acrylic.
Wegood M. Awad +7 more
wiley +1 more source
Cover Time in Edge-Uniform Stochastically-Evolving Graphs
Algorithms, 2018 We define a general model of stochastically-evolving graphs, namely the edge-uniform stochastically-evolving graphs. In this model, each possible edge of an underlying general static graph evolves independently being either alive or dead at each discrete
Ioannis Lamprou +2 more
doaj +1 more source