New algorithm achieves data-dependent regret bounds in MDPs with unknown transitions.
problem Achieving best-of-both-worlds guarantees with data-dependent regret bounds in MDPs with unknown transitions.
method Optimistic follow-the-regularized-leader algorithm with new optimistic Q-function estimators and transition bonus.
result First-order, second-order, and path-length bounds with polylog(T) regret in the stochastic regime.
We present a novel kernel-based machine learning algorithm for identifying the low-dimensional geometry of the effective dynamics of high-dimensional multiscale stochastic systems. Recently, the authors developed a mathematical framework for the computation of optimal reaction coordinates of such systems that is based …
Improved KSD test for better detection of differences in distributions.
problem Low power of KSD test when distributions have same modes but different mixing proportions.
method Perturb the observed sample using Markov transition kernels to improve KSD test power.
result Perturbed KSD test can lead to substantially higher power than the original KSD test.
A recent line of work studies overparametrized neural networks in the "kernel regime," i.e. when the network behaves during training as a kernelized linear predictor, and thus training with gradient descent has the effect of finding the minimum RKHS norm solution. This stands in contrast to other studies which demonstr…
We introduce a new geometric approach that constructs a transition kernel of Markov chain. Our method always minimizes the average rejection rate and even reduce it to zero in many relevant cases, which cannot be achieved by conventional methods, such as the Metropolis-Hastings algorithm or the heat bath algorithm (Gib…
Neural networks with DAGs show linearity as width increases.
problem Understanding linearity in neural networks with arbitrary DAG structures.
method Analyzing the transition to linearity in networks with arbitrary DAGs, characterizing width by minimum in-degree.
result General neural networks with DAGs exhibit linearity as width approaches infinity.
Bayesian Gaussian Processes improve exoplanet transit and Hubble constant inference.
problem Improving exoplanet transit and Hubble constant inference using Bayesian Gaussian Processes.
method Kernel-, mean- and noise-marginalised Gaussian Processes with evidence-based model comparison and transdimensional sampling.
result Inferred Hubble constant H0 values from cosmic chronometers, baryon acoustic oscillations and combined datasets are 66±6kms−1Mpc−1, 67±10kms−1Mpc−1 and 69±6kms−1Mpc−1, respectively. New method detects changes in high-dimensional Markov processes without explicit likelihood evaluation.
problem Quickest change detection in Markov processes with unknown transition kernels.
method Learn conditional score from sample pairs, develop score-based CUSUM procedure.
result Exponential lower bounds on mean time to false alarm and asymptotic upper bounds on detection delay.
Wide neural networks become linear, with constant tangent kernel, due to Hessian scaling.
problem Understanding the linearity of large non-linear models and the tangent kernel.
method Analyzing the scaling properties of the Hessian matrix of neural networks as their width increases.
result The constancy of the tangent kernel is due to the scaling properties of the Hessian matrix.
The existence of stationary Markov perfect equilibria in stochastic games is shown under a general condition called "(decomposable) coarser transition kernels". This result covers various earlier existence results on correlated equilibria, noisy stochastic games, stochastic games with finite actions and state-independe…
Study on how initialization scale affects neural network training regimes.
problem Understanding the transition between kernel and rich regimes in overparametrized models.
method Analysis of simple depth-D models and empirical testing on complex models.
result Scale of initialization controls transition between kernel and rich regimes.
Develops a new framework for conditional independence.
problem Generalizing previous notions of conditional independence.
method Introduces transition probability spaces and transitional random variables.
result Satisfies all desired relevance relations except symmetry.
The paper studies convergence of kernel autocovariance operators for stationary processes.
problem Estimating autocovariance operators of stationary processes on Polish spaces.
method Investigates convergence of empirical estimates of autocovariance operators under various conditions.
result Provides consistency results for kernel PCA and spectral analysis methods.
We consider online learning for minimizing regret in unknown, episodic Markov decision processes (MDPs) with continuous states and actions. We develop variants of the UCRL and posterior sampling algorithms that employ nonparametric Gaussian process priors to generalize across the state and action spaces. When the trans…
A computational technique borrowed from the physical sciences is introduced to obtain accurate closed-form approximations for the transition probability of arbitrary diffusion processes. Within the path integral framework the same technique allows one to obtain remarkably good approximations of the pricing kernels of f…
Kernel-based reinforcement learning (KBRL) stands out among reinforcement learning algorithms for its strong theoretical guarantees. By casting the learning problem as a local kernel approximation, KBRL provides a way of computing a decision policy which is statistically consistent and converges to a unique solution. U…
e-GGPs learn graph vertex transitions over time.
problem Static graph Gaussian Processes cannot handle dynamic graph structures.
method Proposes e-GGPs with a transition function and neighbourhood kernel.
result e-GGPs outperform static GGPs on time-series regression.
Contrastive learning estimates transition kernels for continuous-time stochastic processes.
problem Estimating transition kernels for continuous-time stochastic processes without labeled data.
method Contrastive learning applied to strong-mixing continuous-time stochastic processes.
result Contrastive learning can estimate transition kernels for small-to-mid-range intervals in the diffusion case.
High-dimensional U-statistics show surprising phase transitions, impacting kernel-based tests.
problem Understanding phase transitions in high-dimensional U-statistics.
method Proved a convergence theorem for U-statistics of degree two in high dimensions.
result High-dimensional U-statistics can have non-Gaussian limits with larger variance and asymmetry.
Graph diffusion processes approximate manifold heat semigroups using graph transition matrices.
problem Approximating manifold heat semigroups from graph data under low regularity conditions.
method Iterating graph transition matrix P to approximate Qt=etΔ, bounding error in ∞-norm. result Convergence rates O(N−2/(d+6)) for manifold heat semigroup approximation, valid for in-sample and out-of-sample. New method for optimizing risk in financial models using Fourier transforms.
problem Optimizing risk in financial models with multi-period mean-CVaR.
method Strictly monotone 2D integration scheme via Fourier-trained transition kernels.
result Established robust and accurate optimization method for financial models.
We study optimal solutions to an abstract optimization problem for measures, which is a generalization of classical variational problems in information theory and statistical physics. In the classical problems, information and relative entropy are defined using the Kullback-Leibler divergence, and for this reason optim…
The equivalence of principal bundles with transitive Lie groupoids due to Ehresmann is a well known result. A remarkable generalisation of this equivalence, due to Mackenzie, is the equivalence of principal bundle extensions with those transitive Lie groupoids over the total space of a principal bundle, which also admi…
In this paper we study the local description of spaces of forms on transitive Lie algebroids. We use this local description to introduce global structures like metrics, ∗-Hodge operation and integration along the algebraic part of the transitive Lie algebroid (its kernel). We construct a Čech-de Rham bicomplex wit…
Researchers approximate conditional expectation operators using kernel methods.
problem Statistical approximation of conditional expectation operators under minimal assumptions.
method Modifying the domain of the operator, approximating it by Hilbert-Schmidt operators in a reproducing kernel Hilbert space.
result The nonparametric estimate of the operator converges to a specific limiting object.
We develop algorithms with low regret for learning episodic Markov decision processes based on kernel approximation techniques. The algorithms are based on both the Upper Confidence Bound (UCB) as well as Posterior or Thompson Sampling (PSRL) philosophies, and work in the general setting of continuous state and action …
New algorithm optimizes resource allocation in non-stationary networks.
problem Optimal resource allocation in non-stationary RMABs is computationally hard.
method Sliding-Window Online Whittle (SW-Whittle) policy for non-stationary transition kernels.
result Sub-linear dynamic regret achieved with unknown variation budget.
A new method learns robust policies from offline data with latent structures.
problem Conservative policies under unrealistic dynamics shifts.
method d-RRMDP framework with f-divergence regularization and R2PVI algorithm. result R2PVI learns robust policies with superior computational efficiency.
Kernel Bayesian inference is a principled approach to nonparametric inference in probabilistic graphical models, where probabilistic relationships between variables are learned from data in a nonparametric manner. Various algorithms of kernel Bayesian inference have been developed by combining kernelized basic probabil…
Model place cells as spatial embeddings for efficient path planning and cognitive map construction.
problem Encoding spatial navigation in the hippocampus.
method Model place cells using spectral decomposition of multi-step random walk transition kernels, inducing sparsity and adjacency.
result Place cells encode spatial information through non-negativity and inner-product structure, forming a cognitive map.
Study local convergence of GDA for training GANs with kernel-based discriminators.
problem Analyzing the local dynamics of GDA for GANs with kernel-based discriminators.
method Linearization of a non-linear dynamical system, under an isolated points model assumption.
result Showed phase transitions indicating convergence, oscillation, or divergence of GDA.
Develops a flexible model for regime transitions in time series data.
problem Nonlinear and context-dependent regime transitions in time series data.
method Semi-parametric state-space model with learned transition functions.
result Improved recovery of nonlinear transition dynamics and earlier detection of regime changes.
This work investigates a mixture of LMC and RMHMC with MMALA for geometric ergodicity.
problem Lack of geometric ergodicity study in Riemannian manifold and Lagrangian Monte Carlo methods.
method Investigates a mixture of LMC and RMHMC with MMALA to achieve geometric ergodicity.
result Demonstrates geometric ergodicity in the mixture of LMC and RMHMC with MMALA.
The paper introduces new estimators for multivariate functions using Fourier methods.
problem Estimating multivariate functions like densities and regression functions.
method Monte Carlo estimators based on the Fourier integral theorem.
result Established rates of convergence for new estimators, often superior to existing methods.
Characterizes RFF regression in large n,p,N setting, providing precise learning phases and double descent curve.
problem Characterizes RFF regression in large n,p,N setting. method Characterizes the exact asymptotics of random Fourier feature (RFF) regression in the realistic setting of large n,p,N. result Characterizes two qualitatively different phases of learning and the corresponding double descent test error curve.
WRAAC uses Wasserstein distance for robust reinforcement learning.
problem Lack of quantified robustness to system dynamics in existing reinforcement learning algorithms.
method Leverages Wasserstein distance to connect state disturbance to transition kernel disturbance, reducing infinite-dimensional optimization to a finite-dimensional problem.
result Designs a novel algorithm, WRAAC, that achieves robust reinforcement learning.
In this paper we introduce and study some mathematical structures on top of transitive Lie algebroids in order to formulate gauge theories in terms of generalized connections and their curvature: metrics, Hodge star operator and integration along the algebraic part of the transitive Lie algebroid (its kernel). Explicit…
ARL-GEN adapts to the smallest model class in nested families for RL with improved regret.
problem Model selection for Reinforcement Learning with nested model families.
method Adaptive Reinforcement Learning (ARL-GEN) with value targeted regression and model selection module.
result ARL-GEN achieves a matching regret to an oracle with knowledge of the true model class.
Study reveals how initialization scale affects training accuracy in linear networks.
problem Understanding implicit bias in linear classification models.
method Asymptotic analysis of gradient flow trajectories and training loss minimization.
result Implicit bias is more complex at reasonable initialization scales and training accuracies.
We study the maximum mean discrepancy (MMD) in the context of critical transitions modelled by fast-slow stochastic dynamical systems. We establish a new link between the dynamical theory of critical transitions with the statistical aspects of the MMD. In particular, we show that a formal approximation of the MMD near …
In recent years, non-parametric methods utilizing random walks on graphs have been used to solve a wide range of machine learning problems, but in their simplest form they do not scale well due to the quadratic complexity. In this paper, a new dual-tree based variational approach for approximating the transition matrix…
A multi-task GP model tracks time-varying transition probabilities between two states.
problem Tracking time-varying transition probabilities between 'moves' and 'pauses' states.
method Kernel-based multi-task Gaussian Process model with time-variability and constraints.
result Enforces constraints while learning transition probabilities.
This study analyzes convergence and stability of reinforcement learning algorithms.
problem Understanding the conditions under which reinforcement learning algorithms converge and remain stable.
method Theoretical analysis of convergence and stability of Episodic Upside-Down Reinforcement Learning, Goal-Conditioned Supervised Learning, and Online Decision Transformers.
result The algorithms can achieve near-optimal behavior if the transition kernel is close to a deterministic kernel.
HDT improves MCMC on graphs with history-dependent sampling.
problem Efficient sampling from target distributions on general graphs with low computational overhead.
method History-driven target (HDT) framework that replaces the original target distribution with a history-dependent one.
result Near-zero variance performance and scalability to large graphs with memory-efficient implementation.
Stochastic kernel based dimensionality reduction approaches have become popular in the last decade. The central component of many of these methods is a symmetric kernel that quantifies the vicinity between pairs of data points and a kernel-induced Markov chain on the data. Typically, the Markov chain is fully specified…
Deep networks learn features suddenly, akin to a phase transition.
problem Understanding sudden feature learning in deep networks.
method Adaptive kernel approach applied to two teacher-student models.
result Feature learning and Grokking are analogous to a first-order phase transition.
Paper tackles matrix estimation under arbitrary noise, achieving minimax optimality.
problem Noisy low-rank-plus-sparse matrix recovery under arbitrary dependence.
method Incoherent-constrained least-square estimator, novel energy spreading result.
result Achieves minimax optimality in estimating structured Markov transition kernels.
Quantum theory reinterprets financial pricing by focusing on observable price transitions.
problem Traditional financial models rely on latent variables; this paper proposes a new observable approach.
method Shift operators, spectral calculus, and Lindblad semigroups are used to define observable frequency operators and convolution generators.
result The framework leads to a nonlocal pricing equation that converges to classical Black-Scholes-Merton under small mesh limits.