Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 19 of 19 for “"Bregman divergence"”.
-
Convex regression and its extensions to learning a Bregman divergence and difference of convex functions
… convex regression to learning an arbitrary Bregman divergence and learning a difference of convex functions. We provide approximation guarantees and sample complexity bounds for both these extensions. Furthermore, we provide algorithms to solve the resulting optimization problems based on …
-
Learning vector and functional Bregman divergences for machine learning applications
Functional Bregman divergences are an important class of divergences in machine learning that generalize classical Bregman divergences. Functional Bregman divergences can allow one to define a notion of similarity between both sample points and distributions. This class of divergences includes many …
-
Towards smarter, safer, and swifter generative AI: adapting, guarding, and accelerating diffusion models
… foundation of DMs through the lens of Bregman divergence, extending their applicability to count and range-bounded data domains. We further enhance model intelligence by introducing a dense reward perspective for preference alignment, prioritizing the formative early steps of the …
-
Online EM for Exponential Family Distributions
… descent updates. This method employs custom Bregman divergences for parameter updates and allows for the processing of one unit of data at each iteration. Our second contribution is an improved version of the EM algorithm for exponential family distributions. In this case, we replace the …
-
Geometry and Optimization of Relative Arbitrage
… Part I. It extends the dually flat geometry of Bregman divergence studied by S. Amari and others, leading to plenty of problems for further study.
-
Exploration and Primal-dual Methods in Bandits and Reinforcement Learning
… policy optimization, we propose using dual Bregman divergence instead of the common Euclidean norm to measure similarity in dual space, resulting in the first policy optimization framework with both fast theoretical convergence and superior practical performance. Collectively, these …
-
Scalable clustering algorithms
… a large class of distance measures known as Bregman divergences. We present both hard and soft-clustering algorithms using Bregman divergences. By establishing a bijection between regular exponential family distributions and regular Bregman divergences, we note that Bregman soft clustering …
-
Learning to rank in supervised and unsupervised settings using convexity and monotonicity
… is measured geometrically by a suitable Bregman divergence between the consensus rank and the ranks induced by item specific distributions Real world deployed ranking methods need to be resistant to spam, a particularly sophisticated type of which is link-spam. A popular class of …
-
GEOMETRIC APPROACHES TO DISCRIMINATIVE TRAINING
… of NGD. The loss functions, defined by Bregman divergence, generalizes the log-likelihood objective and can be easily coupled with NGD for optimization. The proposed framework is flexible, allowing us to choose proper convex functions that leads to better training performance. Finally, …
-
Advancing Distributional Off-policy Evaluation in Reinforcement Learning
… (FDE). We have shown that the functional Bregman divergence, which has close relationship with proper scoring rule, serves as an appropriate candidate for building the objective functions. Since proper scoring rule is a widely-studied concept in the realm of statistics, this opens the door …
-
Advancing Distributional Off-policy Evaluation in Reinforcement Learning
… (FDE). We have shown that the functional Bregman divergence, which has close relationship with proper scoring rule, serves as an appropriate candidate for building the objective functions. Since proper scoring rule is a widely-studied concept in the realm of statistics, this opens the door …
-
Advancing Distributional Off-policy Evaluation in Reinforcement Learning
… (FDE). We have shown that the functional Bregman divergence, which has close relationship with proper scoring rule, serves as an appropriate candidate for building the objective functions. Since proper scoring rule is a widely-studied concept in the realm of statistics, this opens the door …
-
Advancing Distributional Off-policy Evaluation in Reinforcement Learning
… (FDE). We have shown that the functional Bregman divergence, which has close relationship with proper scoring rule, serves as an appropriate candidate for building the objective functions. Since proper scoring rule is a widely-studied concept in the realm of statistics, this opens the door …
-
Advancing Distributional Off-policy Evaluation in Reinforcement Learning
… (FDE). We have shown that the functional Bregman divergence, which has close relationship with proper scoring rule, serves as an appropriate candidate for building the objective functions. Since proper scoring rule is a widely-studied concept in the realm of statistics, this opens the door …
-
Advancing Distributional Off-policy Evaluation in Reinforcement Learning
… (FDE). We have shown that the functional Bregman divergence, which has close relationship with proper scoring rule, serves as an appropriate candidate for building the objective functions. Since proper scoring rule is a widely-studied concept in the realm of statistics, this opens the door …
-
A Dynamical Systems Perspective on Optimization Algorithms
… smooth if we quantify convexity in terms of the Bregman divergence, namely $D_f(x,y) \geq \frac{\mu}{p}\|x - y\|^p $, and consider the classes $\mathcal{F}_L$ and $\mathcal{F}_{\mu,L}$ as the edge cases $p\to\infty$ and $p=2$. More generally, introducing the \emph{modulus of convexity} ($\phi$) …
-
Generalized Maximum Entropy, Convexity and Machine Learning
… on deformed logarithms are used to construct Bregman divergences, and together these represent a generalization of relative entropy. The framework is analyzed using convex analysis to charac- terize generalized forms of exponential family distributions. Various connections to the existing …
-
Aggregating credences into a belief
… Moreover, we study distance minimization with Bregman divergence, utility maximization with strict proper scores, and their relationship. Direct heterogeneous belief aggregation rules will also be proposed and studied regarding threshold, distance and epistemic-utility. We provide a new …
-
Structured numerical problems in contemporary applications
… be minimized are the von Neumann and the LogDet Bregman matrix divergences. The algorithm that solves this optimization problem performs matrix updates based on repeated eigendecompositions of diagonal plus rank-one matrices in the case of von Neumann matrix divergence, and Cholesky updates in …