arXiv research

New method tackles reinforcement learning of complex ω-regular objectives without models.

The paper explores how regularization can improve multi-objective learning with high-dimensional data.

Study reveals the regularization effect of variational distributions in VAEs.

Paper proposes ASR framework to improve image generation by incorporating human knowledge.

The study examines MCMC methods for arbitrary objectives and finds likelihood sharpness impacts performance and regularization.

Regularized empirical risk minimization with constrained labels (in contrast to fixed labels) is a remarkably general abstraction of learning. For common loss and regularization functions, this optimization problem assumes the form of a mixed integer program (MIP) whose objective function is non-convex. In this form, t…

We consider a wide range of regularized stochastic minimization problems with two regularization terms, one of which is composed with a linear function. This optimization model abstracts a number of important applications in artificial intelligence and machine learning, such as fused Lasso, fused logistic regression, a…

Study shows convergence of stochastic gradient method for unregularized Wasserstein optimization.

This paper proposes a new optimization objective for value-based deep reinforcement learning. We extend conventional Deep Q-Networks (DQNs) by adding a model-learning component yielding a transcoder network. The prediction errors for the model are included in the basic DQN loss as additional regularizers. This augmente…

G-TRACER optimizes deep learning by promoting flat minima.

SAIL-RevKL improves SAIL's convergence by regularizing the objective function.

New method smooths optimization for sparse regularization.

Proposes a deep Auto-Encoder-like framework for visual-tactile fusion object clustering.

Regularized policies are robust to adversarial rewards.

We explore the question of whether the representations learned by classifiers can be used to enhance the quality of generative models. Our conjecture is that labels correspond to characteristics of natural data which are most salient to humans: identity in faces, objects in images, and utterances in speech. We propose …

Proposes a new method for continual learning in neural networks.

A novel Bayesian optimization framework tackles multi-objective constrained problems.

A new algorithm, Regular Tree Search, tackles non-convex simulation optimization problems.

New method averages SGD iterates to achieve adjustable regularization.

Regularization is a powerful technique for extracting useful information from noisy data. Typically, it is implemented by adding some sort of norm constraint to an objective function and then exactly optimizing the modified objective function. This procedure often leads to optimization problems that are computationally…

Proposes a new regularization technique for neural networks using elliptic operators.

The Lagrangian formalism on a arbitrary non-fibrating manifold is considered. The kinematical description of this generic situation is based on the concept of (higher-order) Grassmann manifolds which is the factorization of the regular velocity manifold to the action of the differential group. Here we introduce in this…

Proposes new stochastic algorithms for multi-objective optimization.

Improves robustness of high-dimensional regression with rank objective and group lasso regularization.

Unified approach to structured prediction combining entropy regularization and neuro-symbolic logic.

Proves hardness of semi-discrete optimal transport and proposes regularization methods.

Proposes a method for multi-view clustering that integrates consistent and complementary graph regularizers.

We propose a formulation of a Lorentzian quantum geometry based on the framework of causal fermion systems. After giving the general definition of causal fermion systems, we deduce space-time as a topological space with an underlying causal structure. Restricting attention to systems of spin dimension two, we derive th…

Study shows policy gradient convergence for entropy-regularized MDPs with neural nets in mean-field regime.

IAGAN method improves medical image reconstruction by incorporating adaptive GAN priors.

Characterizes dropout's regularizer in deep linear networks.

New filter bank regularization improves DCNNs by incorporating image priors.

Path regularization improves GFlowNets exploration and generalization.

Optimizes privacy vs. utility trade-off through optimal transport.

Regularized mixtures improve inflation and interest rate forecasts, especially correcting overconfidence.

There are three principle paradigms of statistical inference: (i) Bayesian, (ii) information-based and (iii) frequentist inference. We describe an objective prior (the weighting or $w$-prior) which unifies objective Bayes and information-based inference. The $w$-prior is chosen to make the marginal probability an unbia…

This paper finds sparsest ReLU networks for interpolating data.

Researchers use estimated Kolmogorov complexity for better link prediction in graphs.

Regularity properties of intrinsic objects for a large class of Stein Manifolds, namely of Monge-Ampère exhaustions and Kobayashi distance, is interpreted in terms of modular data. The results lead to a construction of an infinite dimensional family of convex domains with squared Kobayashi distance of prescribed regula…

Proposes learning regularization strength directly from data.

ProxSPS improves on SPS for regularization tasks, offering better stability and performance.

Develops minibatch stochastic proximal gradient for large-scale learning models.

Theory for soft-margin classifiers on object manifolds.

Sparse Bayesian Optimization (SEBO) finds interpretable configurations.

DAC enhances exploration in reinforcement learning with entropy regularization.

Generative models improve MRI reconstruction by learning image structure.

Paper proposes an algorithm to reduce hypothesis space for faster convergence in high-dimensional settings.

RES, a regularized stochastic version of the Broyden-Fletcher-Goldfarb-Shanno (BFGS) quasi-Newton method is proposed to solve convex optimization problems with stochastic objectives. The use of stochastic gradient descent algorithms is widespread, but the number of iterations required to approximate optimal arguments c…