arXiv research

Aggregator learns optimal forecast aggregation in partial evidence environments.

In this paper, we test a partially segmented ICAPM for two developed markets, two emerging markets and World market, using an asymmetric extension of the multivariate GARCH process of De Santis and Gerard (1997,1998). We find that this asymmetric process provides a significantly better fit of the data than a standard s…

We perform a comparative analysis of the Chinese stock market around the occurrence of the 2008 crisis based on the random matrix analysis of high-frequency stock returns of 1228 stocks listed on the Shanghai and Shenzhen stock exchanges. Both raw correlation matrix and partial correlation matrix with respect to the ma…

Develops a method to infer partial rankings from sparse comparisons.

New particle filter estimates model evidence without bias.

Bayesian neural networks can be partially stochastic without losing predictive power.

Characterizes causal structure dominance for latent variables.

We present a combinatorial proof for the existence of the sign refined grid homology in lens spaces, and a self contained proof that $\partial_{\mathbb{Z}}^2 = 0$. We also present a Sage program that computes $\widehat{\mathrm{GH}} (L(p,q),K;\mathbb{Z})$, and provide empirical evidence supporting the absence of torsion…

In this paper, we examine the problem of approximating a general linear dimensionality reduction (LDR) operator, represented as a matrix $A \in \mathbb{R}^{m \times n}$ with $m < n$, by a partial circulant matrix with rows related by circular shifts. Partial circulant matrices admit fast implementations via Fourier tra…

Study portfolio optimization with partial info and drawdown constraints using deep learning.

DVRL learns a generative model for partially observable environments.

We have analyzed the Indices of Industrial Production (Seasonal Adjustment Index) for a long period of 240 months (January 1988 to December 2007) to develop a deeper understanding of the economic shocks. The angular frequencies estimated using the Hilbert transformation, are almost identical for the 16 industrial secto…

A methodology is developed to identify, as units of study, each decrease in the value of a stock from a given maximum price level. A critical level in the amount of price declines is found to separate a segment operating under a random walk from a segment operating under a power law. This level is interpreted as a poin…

Knot Floer homology detects ribbon concordance.

Optimal algorithms learn Gaussian trees and polytrees from data.

This paper improves RL for PDE control with function-valued actions.

We investigate Riemannian (non-Kahler) Ricci flow solutions that develop finite-time Type-I singularities and present evidence in favor of a conjecture that parabolic rescalings at the singularities converge to singularity models that are shrinking Kahler-Ricci solitons. Specifically, the singularity model for these so…

New method improves generative model performance by fully conditioning variational posteriors.

Bayesian evidence helps compare models but can overfit.

Develops a PIDE framework for option pricing with stochastic volatility and jumps.

Extends PAC learning theory to handle partial concepts with special properties.

We use PDPs with confidence bands to explain HPO results.

Theoretical analysis of MCR for improving imputation quality in partially observed data.

Tighter ELBOs can harm learning, new algorithms improve performance.

Study reveals latent state computation in stochastic volatility models.

SigGPDE scales sparse Gaussian processes for sequential data.

Study community detection in multi-view data with various types of information.

Active inference implemented for high-dimensional tasks shows efficient exploration and improved sample efficiency.

This study uses quantile regression to analyze U.S. firms' capital structure across different leverage levels.

We propose Rademacher complexity bounds for multiclass classifiers trained with a two-step semi-supervised model. In the first step, the algorithm partitions the partially labeled data and then identifies dense clusters containing $κ$ predominant classes using the labeled training examples such that the proportion of t…

New statistical properties for mini-batch Cox-NN optimization.

Photography method solves manifold invariants.

Coordinate descent methods employ random partial updates of decision variables in order to solve huge-scale convex optimization problems. In this work, we introduce new adaptive rules for the random selection of their updates. By adaptive, we mean that our selection rules are based on the dual residual or the primal-du…

New method for robust matrix completion with mixed data types.

We introduce a stochastic price model where, together with a random component, a moving average of logarithmic prices contributes to the price formation. Our model is tested against financial datasets, showing an extremely good agreement with them. It suggests how to construct trading strategies which imply a capital g…

HRM-Agent learns to navigate dynamic mazes using reinforcement learning.

The paper divides minimal surfaces in a ball into two parts.

This paper improves system identification by reducing sample complexity for high-dimensional linear dynamical systems.

The paper explores how to handle uncertain evidence in probabilistic models.

GEAR uses graphs to integrate and reason over multiple evidence for fact verification.

The paper studies inequalities for fractional GJMS operators on conformal infinity.

New machine learning model identifies key drivers of market troughs.

Regulated Bitcoin futures led to higher volatility and trading volume.

Evidence acquisition costs influence disclosure behavior and preference.

New method identifies flawed internal models of the world in animals.

Enhances clustering by using external categorical data.

This paper evaluates a method to improve representations using incomplete external evidence across tasks.

This article introduces a framework to estimate the value of evidence-based decision making.