arXiv research

Bottom-up algorithms outperform top-down in hierarchical community detection at intermediate levels.

New research shows fairness in machine learning can sometimes make disadvantaged groups worse off.

In the top-down approach to multi-name credit modeling, calculation of singe name sensitivities appears possible, at least in principle, within the so-called random thinning (RT) procedure which dissects the portfolio risk into individual contributions. We make an attempt to construct a practical RT framework that enab…

Large-scale classification of data where classes are structurally organized in a hierarchy is an important area of research. Top-down approaches that exploit the hierarchy during the learning and prediction phase are efficient for large scale hierarchical classification. However, accuracy of top-down approaches is poor…

Paper uses LLMs for sector allocation, showing better returns.

Efficient algorithm for evaluating hierarchical classification methods at multiple operating points.

A new method for feature fusion in U-Net decoders using difference-based gating.

In this paper, we propose a novel generative model named Stacked Generative Adversarial Networks (SGAN), which is trained to invert the hierarchical representations of a bottom-up discriminative network. Our model consists of a top-down stack of GANs, each learned to generate lower-level representations conditioned on …

A new multi-phase approach improves supply chain forecasting accuracy.

CNT leverages noisy targets to guide model learning.

Explainable Artificial Intelligence (XAI)has received a great deal of attention recently. Explainability is being presented as a remedy for the distrust of complex and opaque models. Model agnostic methods such as LIME, SHAP, or Break Down promise instance-level interpretability for any complex machine learning model. …

Model learns collective and individual dynamics in time series data.

Method generates joint posterior samples of source and foreground mass distributions for gravitational lensing.

Reduces high granularity and dimensionality in hierarchical categorical variables.

Director learns hierarchical behaviors from pixels, outperforming exploration methods.

The paper proves Gorenstein contractions for multiscale differentials on nodal curves.

In this paper we study the classification of ancient convex solutions to the mean curvature flow in $\R^{n+1}$. An open problem related to the classification of type II singularities is whether a convex translating solution is $k$-rotationally symmetric for some integer $2\le k\le n$, namely whether its level set is a …

The paper analyzes insurance risk with Parisian ruin and capital injection.

Criterion found for blowing down in 6D symplectic geometry.

Study reveals significant performance flips in GLOD using repurposed graph classification datasets.

A new hierarchical forecasting method improves overall accuracy.

In this study, we investigate the use of global information to speed up the learning process and increase the cumulative rewards of reinforcement learning (RL) in competition tasks. Within the actor-critic RL, we introduce multiple cooperative critics from two levels of the hierarchy and propose a reinforcement learnin…

Paper presents a robust transfer learning method for active level set estimation.

Analytical, free of time consuming Monte Carlo simulations, framework for credit portfolio systematic risk metrics calculations is presented. Techniques are described that allow calculation of portfolio-level systematic risk measures (standard deviation, VaR and Expected Shortfall) as well as allocation of risk down to…

The quality of data representation in deep learning methods is directly related to the prior model imposed on the representations; however, generally used fixed priors are not capable of adjusting to the context in the data. To address this issue, we propose deep predictive coding networks, a hierarchical generative mo…

Firms delay write-downs for adverse macroeconomic and industry outcomes but not for firm-specific issues.

Hierarchical forecasting (HF) is needed in many situations in the supply chain (SC) because managers often need different levels of forecasts at different levels of SC to make a decision. Top-Down (TD), Bottom-Up (BU) and Optimal Combination (COM) are common HF models. These approaches are static and often ignore the d…

Analytical, free of time consuming Monte Carlo simulations, framework for credit portfolio systematic risk metrics calculations is presented. Techniques are described that allow calculation of portfolio-level systematic risk measures (standard deviation, VaR and Expected Shortfall) as well as allocation of risk down to…

In this paper a novel modification of the multilevel Monte Carlo approach, allowing for further significant complexity reduction, is proposed. The idea of the modification is to use the method of control variates to reduce variance at level zero. We show that, under a proper choice of control variates, one can reduce t…

The paper breaks down AUC into cluster-level components for better model diagnostics.

Gonogo offers tools for sensitivity experiments in R.

The paper analyzes orbits of integer tuples using braid diagrams.

This paper studies how label noise affects Federated Learning.

In this paper it is shown that the space of tight geodesic segments connecting any two vertices in a complex of cycles has finite, uniformly bounded dimension. The dimension is defined in terms of a discrete analogue of Jacobi fields, which are explicitly constructed and shown to give a complete description of the enti…

ESOP uses Bayesian optimization to find optimal lock-down schedules.

Attention mechanism combines bottom-up and top-down signals in neural networks.

We compare the star surgery operations introduced in [KS] to the generalized rational blow-down. We show that star surgery shares the properties that make rational blow-down useful for constructions of small exotic symplectic 4-manifolds. Then we show that star surgery operations provide a strictly more general class o…

Compositional structures between parts and objects are inherent in natural scenes. Modeling such compositional hierarchies via unsupervised learning can bring various benefits such as interpretability and transferability, which are important in many downstream tasks. In this paper, we propose the first deep latent vari…

A generative Bayesian model is developed for deep (multi-layer) convolutional dictionary learning. A novel probabilistic pooling operation is integrated into the deep model, yielding efficient bottom-up and top-down probabilistic learning. After learning the deep convolutional dictionary, testing is implemented via dec…

The blow-down map is studied in Lie algebroid cohomology.

We present the nested Chinese restaurant process (nCRP), a stochastic process which assigns probability distributions to infinitely-deep, infinitely-branching trees. We show how this stochastic process can be used as a prior distribution in a Bayesian nonparametric model of document collections. Specifically, we presen…

2020 Census uses more noise to protect privacy than needed, improving data accuracy.

New surfaces with conjugate points have global blow-down maps in their TT spaces.

HHAR-net uses neural networks to recognize human activities at different levels of abstraction.

Study ancient solutions on noncompact steady Ricci solitons, proving types of ancient solutions.

Future predictions on sequence data (e.g., videos or audios) require the algorithms to capture non-Markovian and compositional properties of high-level semantics. Context-free grammars are natural choices to capture such properties, but traditional grammar parsers (e.g., Earley parser) only take symbolic sentences as i…

For machine learning perception problems, human-level classification performance is used as an estimate of top algorithm performance. Thus, it is important to understand as precisely as possible the factors that impact human-level performance. Knowing this 1) provides a benchmark for model performance, 2) tells a proje…

This paper presents the first theoretical results showing that stable identification of overcomplete $μ$-coherent dictionaries $Φ\in \mathbb{R}^{d\times K}$ is locally possible from training signals with sparsity levels $S$ up to the order $O(μ^{-2})$ and signal to noise ratios up to $O(\sqrt{d})$. In particular the di…