arXiv research

Flexible evidential deep learning improves uncertainty quantification in machine learning.

RegFlow models future states with flexible probability distributions.

LEMs extend transformer-based architectures for complex execution problems.

New method for robust financial portfolio analysis.

Recent developments in machine-learning algorithms have led to impressive performance increases in many traditional application scenarios of artificial intelligence research. In the area of deep reinforcement learning, deep learning functional architectures are combined with incremental learning schemes for sequential …

New method detects changes online with bounds on delay.

Estimates outcomes under hypothetical scenarios using a flexible framework.

This study provides benchmarks for different implementations of LSTM units between the deep learning frameworks PyTorch, TensorFlow, Lasagne and Keras. The comparison includes cuDNN LSTMs, fused LSTM variants and less optimized, but more flexible LSTM implementations. The benchmarks reflect two typical scenarios for au…

DINo forecasts PDEs with flexible extrapolation and adaptability.

This paper proposes an agent-based model that combines both spot and balancing electricity markets. From this model, we develop a multi-agent simulation to study the integration of the consumers' flexibility into the system. Our study identifies the conditions that real-time prices may lead to higher electricity costs,…

Agent-based models have demonstrated their power and flexibility in Econophysics. However their major challenge is still to devise more realistic simulation scenarios. The complexity of Economy makes appealing the idea of introducing chaotic number generators as simulation engines in these models. Chaos based number ge…

Transformer model handles causal inference with DAG integration.

Generative Networks outperform traditional methods in PiT ESG generation.

By interpreting a traffic scene as a graph of interacting vehicles, we gain a flexible abstract representation which allows us to apply Graph Neural Network (GNN) models for traffic prediction. These naturally take interaction between traffic participants into account while being computationally efficient and providing…

Simformer uses transformer models to perform flexible Bayesian inference.

Novel framework for reliable long-tailed classification.

A new method combines experts' opinions to train regression models with noisy labels.

Flexible empirical Bayes for large-scale multiple linear regression.

In this paper, we present a two-stage stochastic international portfolio optimisation model to find an optimal allocation for the combination of both assets and currency hedging positions. Our optimisation model allows a "currency overlay", or a deviation of currency exposure from asset exposure, to provide flexibility…

New method tackles high-dimensional contextual bandits with flexible kernel models.

Method enhances anomaly detection using contrastive learning and out-of-distribution data.

Gaussian processes (GPs) are commonplace in spatial statistics. Although many non-stationary models have been developed, there is arguably a lack of flexibility compared to equipping each location with its own parameters. However, the latter suffers from intractable computation and can lead to overfitting. Taking the i…

Proposes a framework for modeling RTB auctions using point processes.

Unified framework for N-tuples learning improves weakly supervised tasks.

Proposes a flexible method for learning latent causal representations.

Humans and animals show remarkable flexibility in adjusting their behaviour when their goals, or rewards in the environment change. While such flexibility is a hallmark of intelligent behaviour, these multi-task scenarios remain an important challenge for machine learning algorithms and neurobiological models alike. We…

New methods generalize nonlinear ICA beyond structural sparsity.

Proposes a new allocation method for distributionally robust ranking and selection.

New framework using Jensen-Shannon divergence improves domain adaptation theory.

CDST improves ensemble prediction by adjusting model weights based on covariates.

In this paper, we introduce a novel, non-recursive, maximal matching algorithm for double auctions, which aims to maximize the amount of commodities to be traded. It differs from the usual equilibrium matching, which clears a market at the equilibrium price. We compare the two algorithms through experimental analyses, …

We propose a unified methodology to input non-linear views from any number of users in fully general non-normal markets, and perform, among others, stress-testing, scenario analysis, and ranking allocation. We walk the reader through the theory and we detail an extremely efficient algorithm to easily implement this met…

Bayesian Gaussian Process ODEs enhanced with normalizing flows for improved flexibility and accuracy.

Paper proposes a robust hypothesis testing method using Sinkhorn distance.

Improves robustness of information bottleneck framework with sparsity-inducing prior.

Tensor models improve joint EEG and fMRI analysis.

In large scale systems, approximate nearest neighbour search is a crucial algorithm to enable efficient data retrievals. Recently, deep learning-based hashing algorithms have been proposed as a promising paradigm to enable data dependent schemes. Often their efficacy is only demonstrated on data sets with fixed, limite…

Proposes a flexible feature allocation model for sparse factor analysis.

CausalCompass evaluates TSCD robustness under violations of modeling assumptions.

Unified approach for non-stationary and clustered bandits.

Secure Multiparty Computation protects data privacy in Symbolic Regression.

pmsims R package uses Gaussian process for flexible sample size estimation in clinical models.

Improved nested simulation for financial risk measurement.

Enhancing spectral embedding for low-dimensional embeddings in rare disease cohorts

Adaptive batch sizes improve active learning efficiency and flexibility.

Proposes a flexible neural recommendation framework for better prediction performance.

Developed a flexible Bayesian g-formula for causal survival analysis with time-dependent confounding.

Generative Score Inference improves uncertainty quantification for multimodal data.