PDERL improves evolutionary reinforcement learning by using learning-based variation operators.
problem Scalability issue in Genetic Algorithms when combined with Deep Neural Networks.
method Integrates evolutionary and reinforcement learning through a hierarchical approach with learning-based variation operators.
result PDERL outperforms traditional evolutionary and reinforcement learning methods in robot locomotion tasks.
We propose a deep learning based method, the Deep Ritz Method, for numerically solving variational problems, particularly the ones that arise from partial differential equations. The Deep Ritz method is naturally nonlinear, naturally adaptive and has the potential to work in rather high dimensions. The framework is qui…
A modular GP framework for efficient transfer learning.
problem Efficiently transfer knowledge across different tasks or datasets.
method Modular variational Gaussian processes (GPs) with a dictionary of well-fitted GPs.
result Reduces computational costs and allows the transfer of uncertainty metrics.
With the growing prevalence of smart grid technology, short-term load forecasting (STLF) becomes particularly important in power system operations. There is a large collection of methods developed for STLF, but selecting a suitable method under varying conditions is still challenging. This paper develops a novel reinfo…
Variational inference is an umbrella term for algorithms which cast Bayesian inference as optimization. Classically, variational inference uses the Kullback-Leibler divergence to define the optimization. Though this divergence has been widely used, the resultant posterior approximation can suffer from undesirable stati…
We derive the first and second variation formula for the Green's function pole's value of Paneitz operator on the standard three sphere. In particular it is shown that the first variation vanishes and the second variation is nonpositively definite. Moreover, the second variation vanishes only at the direction of confor…
New algorithm for RL using mean embeddings of return distributions.
problem Improving reinforcement learning algorithms for dynamic programming.
method Mean embeddings of return distributions, novel algorithms for RL.
result Asymptotic convergence and improved performance in deep RL.
Anomaly detection scores from VAE gradients improve tumor detection.
problem Improving anomaly detection in medical imaging.
method Using Variational Autoencoders to approximate anomaly ratings.
result Variance Autoencoder gradient-based ratings outperform other methods in tumor detection.
We develop a probabilistic framework for deep learning based on the Deep Rendering Mixture Model (DRMM), a new generative probabilistic model that explicitly capture variations in data due to latent task nuisance variables. We demonstrate that max-sum inference in the DRMM yields an algorithm that exactly reproduces th…
A new algorithm speeds up matrix multiplication without actual multiplication.
problem Efficiently multiplying matrices in machine learning.
method Learning-based algorithm that uses hashing, averaging, and byte shuffling.
result Often runs 100x faster than exact matrix products and 10x faster than current approximate methods.
A new SL strategy optimizes HVAC DR in multi-zone buildings.
problem Optimal DR of HVAC units in multi-zone buildings is challenging.
method Supervised learning with ANN replication and DNN integration.
result SLAMP achieves effective DR schedules with reduced computation time.
Study uses machine learning to solve photoacoustic tomography's inverse problem.
problem Solving the full inverse problem in photoacoustic tomography.
method Developed an approach using variational autoencoders for Bayesian estimation of the posterior distribution.
result Evaluated the approach with numerical simulations and compared it to a Bayesian solution.
VANO uses neural operators for unsupervised learning of functional data.
problem Learning operators between infinite dimensional spaces for functional data.
method Variational Autoencoding Neural Operators (VANO) approach.
result VANO can learn and reconstruct functional data without supervision.
A strategy for spectrum sharing in CRNs with multiple PT power levels.
problem Efficient spectrum usage for secondary users in CRNs with multiple PT power levels.
method Data-driven/machine learning based multi-level spectrum sensing and prediction-transmission structures.
result The proposed strategy effectively aligns the ST with the PT power levels, improving spectrum usage.
Proposes D-LADMM for solving constrained optimization problems.
problem Constrained optimization problems with ill-posed inverse problems.
method Introduces Differentiable Linearized ADMM (D-LADMM) with learnable weights and activation functions.
result Rigorously proves globally converged solutions for D-LADMM.
Study efficient neural operator learning using variation spaces.
problem Operator learning using encoder-decoder neural networks.
method Introduce variation space for nonlinear operators, establish approximation bounds.
result Algebraic approximation and learning rates for polynomially decaying input and output encoding errors.
This paper bridges statistical and machine learning approaches to variational inference.
problem Statisticians struggle to understand variational inference from a Frequentist perspective.
method Explains VI, VAEs, and DDMs from a Frequentist viewpoint, starting with EM.
result VI emerges as a scalable solution for intractable E-steps in VAEs and DDMs.
For a scalar evolution equation ut=K(t,x,u,ux,…,un),n≥2 the cohomology spaces H1,s(R∞) vanishes for s≥3 while the space H1,2(R∞) is isomorphic to the space of variational operators. The cohomology space H1,2(R∞) is also shown to be …
Anomaly detection using dimensionality reduction has been an essential technique for monitoring multidimensional data. Although deep learning-based methods have been well studied for their remarkable detection performance, their interpretability is still a problem. In this paper, we propose a novel algorithm for estima…
New kernels allow learning from non-separable data.
problem Learning from non-separable data.
method Introducing entangled kernels and a two-step algorithm.
result Efficient algorithm for learning entangled kernels.
The extragradient method fails for hypomonotone variational inequalities.
problem The convergence of the extragradient method for hypomonotone variational inequalities.
method Application of the extragradient method to hypomonotone linear operators.
result The extragradient method diverges for hypomonotone variational inequalities.
D-GAN predicts spatio-temporal data without explicit factor listing.
problem Challenges in predicting spatio-temporal data due to complexity, variability, and external factors.
method D-GAN uses a deep generative adversarial network to learn spatio-temporal correlations and variations implicitly.
result D-GAN outperforms traditional and deep learning methods in spatio-temporal prediction accuracy.
GeoFunFlow tackles inverse problems on complex geometries with efficient learning.
problem Challenges in inverse problems governed by PDEs, especially on irregular geometries.
method Combines geometric function autoencoder and latent diffusion model trained via rectified flow.
result Achieves state-of-the-art reconstruction accuracy and efficient inference.
New method identifies vanishing arcs for curve singularities.
problem Characterizing arcs sent to geometric vanishing cycles.
method Introducing geometric variation operator and vanishing arcsets.
result Existence of topological exceptional collections of arcsets.
We establish higher-order weighted Sobolev and Holder regularity for solutions to variational equations defined by the elliptic Heston operator, a linear second-order degenerate-elliptic operator arising in mathematical finance. Furthermore, given C∞-smooth data, we prove C∞-regularity of solutions up t…
This paper learns variational models and solvers for inverse problems from incomplete data.
problem Solving inverse problems with partially observed data.
method Joint learning of variational cost and gradient-based solver as neural networks.
result Joint learning leads to improved reconstruction performance.
Study on eigenvalues of complex Hessian operator on pseudoconvex manifolds.
problem Eigenvalue problem for complex Hessian operator on pseudoconvex manifolds.
method Established C1,1-regularity and uniqueness of the first eigenfunction, derived variational formula for the first eigenvalue. result Derivation of a bifurcation-type theorem and geometric bounds for the eigenvalue.
New algorithm reduces training time for deep learning in financial hedging.
problem Optimal hedging in markets with transaction costs.
method ST-Hedging algorithm combining deep learning and FBSDE solver.
result Achieves state-of-the-art performance and scalability.
The paper explores how control variates can reduce variance in Monte Carlo simulations, especially for Sobolev functions.
problem Efficiency of control variates in reducing variance for Monte Carlo simulations.
method Study of a specific quadrature rule using nonparametric regression-adjusted control variates.
result A specific quadrature rule can improve the Monte Carlo rate and achieve the minimax optimal rate under sufficient smoothness assumptions.
New deep learning method solves TSP faster and more efficiently.
problem Approximately solving the Travelling Salesman Problem on 2D Euclidean graphs.
method Uses Graph Convolutional Networks for efficient TSP graph representations and non-autoregressive beam search.
result Significantly reduces optimality gap for large problem instances.
Harmonic gauge simplifies geometric analysis of Riemannian metrics.
problem Analyzing the Hilbert-Einstein functional and its stability.
method Developed a harmonic gauge to eliminate divergence terms and induce elliptic structure.
result Positivity of curvature operator implies spectral stability of the functional.
Study fourth-order geometric flow of shape operator for co-dimension one immersions.
problem Analyzing the geometry of isometric immersions in Riemannian manifolds.
method Introduce a moduli flow to decrease curvature variation energy.
result The flow decreases a natural energy measuring curvature variation.
Enhanced DeepONet framework with uncertainty quantification for complex operators.
problem Learning complex operators with uncertainty quantification.
method Generalised variational inference (GVI) using Rényi's α-divergence.
result Superior predictive accuracy and uncertainty quantification.
Unified deep learning framework solves various optimal transport problems.
problem Solving variational problems in optimal transport with computational challenges.
method Unified deep learning framework leveraging dual formulation of Lagrangians.
result Outperforms previous approaches in single-cell trajectory inference.
Variational methods yield formulas for eigenvalues of elliptic operators, with applications to metric evolution.
problem Deriving formulas for eigenvalues of elliptic operators on compact manifolds.
method Variational methods applied to elliptic operators on compact Riemannian manifolds.
result Generic subsets of metrics yield simple spectra of elliptic operators.
In this paper we provide a detailed proof of the second variation formula, essentially due to Richard Hamilton, Tom Ilmanen and the first author, for Perelman's ν-entropy. In particular, we correct an error in the stability operator stated in Theorem 6.3 of [2]. Moreover, we obtain a necessary condition for linearly …
The Heston stochastic volatility process, which is widely used as an asset price model in mathematical finance, is a paradigm for a degenerate diffusion process where the degeneracy in the diffusion coefficient is proportional to the square root of the distance to the boundary of the half-plane. The generator of this p…
InClass nets use neural networks to estimate CIMMs without assuming fixed parameters.
problem Nonparametric estimation of conditional independence mixture models.
method Independent classifier neural networks (NNs) for multi-class classification.
result Nonparametric identifiability conditions for bivariate CIMMs.
Unsupervised learning can leverage large-scale data sources without the need for annotations. In this context, deep learning-based auto encoders have shown great potential in detecting anomalies in medical images. However, state-of-the-art anomaly scores are still based on the reconstruction error, which lacks in two e…
New method accelerates energetic variational inference using particle dynamics.
problem Efficiently solving variational inference problems with reduced computational cost.
method Particle-based variational inference with implicit scheme, inspired by energy quadratization and operator splitting.
result Significantly reduces computational cost compared to existing methods.
Improved sampling via learned diffusions using variational losses.
problem Sampling from target distributions without direct access to samples.
method Generalized Schrödinger bridge problem, variational formulation, gradient-based optimization.
result Proposed log-variance loss leads to improved performance.
Constructs Lepage equivalents for arbitrary-order Lagrangians.
problem Creating Lepage equivalents for complex Lagrangians.
method Uses variational bicomplex and symmetric linear connections to construct Lepage equivalents satisfying the closure property.
result Shows how to extend global Lepage equivalents to ones satisfying the closure property.
We introduce a learning-based framework to optimize tensor programs for deep learning workloads. Efficient implementations of tensor operators, such as matrix multiplication and high dimensional convolution, are key enablers of effective deep learning systems. However, existing systems rely on manually optimized librar…
A new method solves variational inequality problems with multiple constraints without needing optimal Lagrange multipliers.
problem Solving variational inequality problems with multiple functional constraints efficiently.
method Constrained Gradient Method (CGM) for Minty variational inequality problems.
result The Constrained Gradient Method achieves complexity similar to projection-based methods but with cheaper oracles.
We first generalize the operation of formal exterior differential in the case of finite dimensional fibered manifolds and then we extend it to certain bundles of smooth maps. In order to characterize the operator order of some morphisms between our bundles of smooth maps, we introduce the concept of fiberwise (k,r)-j…
Develops inference combinators for probabilistic programs using neural networks.
problem Creating efficient proposals for probabilistic program inference.
method Inference combinators using neural network parameterization of proposals.
result Correct by construction variational methods tailored to specific models.
New Stein operator improves robustness in model inference.
problem Improving robustness in inference for unnormalized models.
method Density-power weighted Stein operator (γ-Stein operator). result Robust methods for goodness-of-fit testing and posterior approximation.
Defines a filtration on variational bicomplex for concise functional form conditions.
problem Expressing functional form vanishing conditions concisely.
method Introduces a filtration on the variational bicomplex and studies its properties.
result Graded components of the filtration inherit module structures, simplifying functional form conditions.