arXiv research

SMEs provide a transparent testbed for RL evaluation.

Algorithm creates synthetic experiences to enhance Deep Reinforcement Learning.

The paper proposes a machine learning framework for portfolio optimization with limited data.

New algorithm minimizes FE objectives for synthetic AIF agents.

Synthetic augmentation improves financial machine learning performance in variance-dominant regimes.

Paper generates synthetic radar signatures for motion classification.

ICYM2I corrects missingness bias in multimodal learning.

We propose a lifelong learning architecture, the Neural Computer Agent (NCA), where a Reinforcement Learning agent is paired with a predictive model of the environment learned by a Differentiable Neural Computer (DNC). The agent and DNC model are trained in conjunction iteratively. The agent improves its policy in simu…

Two environments are enough to infer causal graphs and counterfactuals.

This paper presents a methodology that aims at the incremental representation of areas inside environments in terms of attractive forces. It is proposed a parametric representation of velocity fields ruling the dynamics of moving agents. It is assumed that attractive spots in the environment are responsible for modifyi…

Vision and learning have made significant progress that could improve robotics policies for complex tasks and environments. Learning deep neural networks for image understanding, however, requires large amounts of domain-specific visual data. While collecting such data from real robots is possible, such an approach lim…

This research generates synthetic data streams for handling concept drifts and novel classes.

Synthetic aperture sonar (SAS) imagery can generate high resolution images of the seafloor. Thus, segmentation algorithms can be used to partition the images into different seafloor environments. In this paper, we compare two possibilistic segmentation approaches. Possibilistic approaches allow for the ability to detec…

Proposes efficient calibration for indoor localization models.

New algorithm for nonstationary multi-armed bandits with optimal performance.

Bayesian approach for learning from synthetic data, improving model accuracy.

Multi-armed bandit algorithms have become a reference solution for handling the explore/exploit dilemma in recommender systems, and many other important real-world problems, such as display advertisement. However, such algorithms usually assume a stationary reward distribution, which hardly holds in practice as users' …

This dissertation automates deep learning pipelines and uses meta-learning for better model selection and data augmentation.

We develop the first general semi-bandit algorithm that simultaneously achieves $\mathcal{O}(\log T)$ regret for stochastic environments and $\mathcal{O}(\sqrt{T})$ regret for adversarial environments without knowledge of the regime or the number of rounds $T$. The leading problem-dependent constants of our bounds are …

Detect hidden confounding in observational data using multiple environments.

Augment small datasets with synthetic backgrounds to train lightweight CNNs for human pose estimation.

Study finds reinforcement learning performance plateaus due to environmental interference.

Kernel-based methods exhibit well-documented performance in various nonlinear learning tasks. Most of them rely on a preselected kernel, whose prudent choice presumes task-specific prior information. Especially when the latter is not available, multi-kernel learning has gained popularity thanks to its flexibility in ch…

ISL improves causal structure learning with invariant structures across different environments.

Algorithm identifies and transfers unstable features to create robust classifiers.

Efficiently adapting to new environments and changes in dynamics is critical for agents to successfully operate in the real world. Reinforcement learning (RL) based approaches typically rely on external reward feedback for adaptation. However, in many scenarios this reward signal might not be readily available for the …

FinStressTS creates synthetic benchmarks for financial forecasting, revealing model weaknesses.

Generative Policy-based Models aim to enable a coalition of systems, be they devices or services to adapt according to contextual changes such as environmental factors, user preferences and different tasks whilst adhering to various constraints and regulations as directed by a managing party or the collective vision of…

Investigates safe decision-making in interactive environments.

Study generates synthetic fNIRS data and applies machine learning for improved neuroimaging.

Audit financial machine learning workflows to detect spurious predictability.

New method ISR improves domain generalization with provable guarantees.

KRCD detects unobserved confounders in nonlinear observational data.

DAL enhances black-box bandit algorithms for non-stationary environments.

TSGM generates synthetic time series data for ML research.

Synthetic Petri Dish predicts neural architecture performance faster.

New algorithms identify invariant features for domain generalization.

HealthSyn generates synthetic user behavior data for health interventions.

In machine learning, the one-class classification problem occurs when training instances are only available from one class. It has been observed that making use of this class's structure, or its different contexts, may improve one-class classifier performance. Although this observation has been demonstrated for static …

Algorithm detects unmeasured confounding in observational data.

New method optimizes policies in non-stationary environments.

This work studies the problem of batch off-policy evaluation for Reinforcement Learning in partially observable environments. Off-policy evaluation under partial observability is inherently prone to bias, with risk of arbitrarily large errors. We define the problem of off-policy evaluation for Partially Observable Mark…

Concept drift is formally defined as the change in joint distribution of a set of input variables X and a target variable y. The two types of drift that are extensively studied are real drift and virtual drift where the former is the change in posterior probabilities p(y|X) while the latter is the change in distributio…

A new memory system handles non-stationary environments by self-sizing and retaining memories.

Highly Autonomous Driving (HAD) systems rely on deep neural networks for the visual perception of the driving environment. Such networks are trained on large manually annotated databases. In this work, a semi-parametric approach to one-shot learning is proposed, with the aim of bypassing the manual annotation step requ…

Stable health predictions need deconfounding test set features.

In data stream mining, predictive models typically suffer drops in predictive performance due to concept drift. As enough data representing the new concept must be collected for the new concept to be well learnt, the predictive performance of existing models usually takes some time to recover from concept drift. To spe…

New bandit algorithm detects and adapts to seasonal changes in rewards.