Although the ongoing success of deep learning is remarkable, the increasing data, model and training algorithm complexity makeathorough understanding oftheir inner workings increasingly difficult.

Identifying unfamiliar inputs, also known as out-of-distribution (OOD) detection, is a crucial property of any decision making process. A simple and empirically validated technique is based on deep ensembles where the variance of predictions over different neural networks acts as a substitute for input uncertainty. Nevertheless, a theoretical understanding of the inductive biases leading to the performance of deep ensemble's uncertainty estimation is missing. To improve our description of their behavior, we study deep ensembles with large layer widths operating in simplified linear training regimes, in which the functions trained with gradient descent can be described by the neural tangent kernel. We identify two sources of noise, each inducing a distinct inductive bias in the predictive variance at initialization. We further show theoretically and empirically that both noise sources affect the predictive variance of non-linear deep ensembles in toy models and realistic settings after training. Finally, we propose practical ways to eliminate part of these noise sources leading to significant changes and improved OOD detection in trained deep ensembles.

We show that Deep Neural Networks introduce two heteroscedastic Gumbel noise sources into Q-Learning. To account for these noise sources, we propose Double Gumbel Q-Learning, a Deep Q-Learning algorithm applicable for both discrete and continuous control. In discrete control, we derive a closed-form expression for the loss function of our algorithm. In continuous control, this loss function is intractable and we therefore derive an approximation with a hyperparameter whose value regulates pessimism in Q-Learning. We present a default value for our pessimism hyperparameter that enables DoubleGum to outperform DDPG, TD3, SAC, XQL, quantile regression, and Mixture-of-Gaussian Critics in aggregate over 33 tasks from DeepMind Control, MuJoCo, MetaWorld, and Box2D and show that tuning this hyperparameter may further improve sample efficiency.

Deep learning models are increasingly used for radiographic analysis, but their reliability is challenged by the stochastic noise inherent in clinical imaging. A systematic, cross-task understanding of how different noise types impact these models is lacking. Here, we evaluate the robustness of state-of-the-art convolutional neural networks (CNNs) to simulated quantum (Poisson) and electronic (Gaussian) noise in two key chest X-ray tasks: semantic segmentation and pulmonary disease classification. Using a novel, scalable noise injection framework, we applied controlled, clinically-motivated noise severities to common architectures (UNet, DeepLabV3, FPN; ResNet, DenseNet, EfficientNet) on public datasets (Landmark, ChestX-ray14). Our results reveal a stark dichotomy in task robustness. Semantic segmentation models proved highly vulnerable, with lung segmentation performance collapsing under severe electronic noise (Dice Similarity Coefficient drop of 0.843), signifying a near-total model failure. In contrast, classification tasks demonstrated greater overall resilience, but this robustness was not uniform. We discovered a differential vulnerability: certain tasks, such as distinguishing Pneumothorax from Atelectasis, failed catastrophically under quantum noise (AUROC drop of 0.355), while others were more susceptible to electronic noise. These findings demonstrate that while classification models possess a degree of inherent robustness, pixel-level segmentation tasks are far more brittle. The task- and noise-specific nature of model failure underscores the critical need for targeted validation and mitigation strategies before the safe clinical deployment of diagnostic AI.

Principled mechanisms for uncertainty quantification would rely on Bayesian inference with an appropriate prior.

Identifying unfamiliar inputs, also known as out-of-distribution (OOD) detection, is a crucial property of any decision making process. A simple and empirically validated technique is based on deep ensembles where the variance of predictions over different neural networks acts as a substitute for input uncertainty. Nevertheless, a theoretical understanding of the inductive biases leading to the performance of deep ensemble's uncertainty estimation is missing. To improve our description of their behavior, we study deep ensembles with large layer widths operating in simplified linear training regimes, in which the functions trained with gradient descent can be described by the neural tangent kernel. We identify two sources of noise, each inducing a distinct inductive bias in the predictive variance at initialization. We further show theoretically and empirically that both noise sources affect the predictive variance of non-linear deep ensembles in toy models and realistic settings after training.

We show that Deep Neural Networks introduce two heteroscedastic Gumbel noise sources into Q-Learning. To account for these noise sources, we propose Double Gumbel Q-Learning, a Deep Q-Learning algorithm applicable for both discrete and continuous control. In discrete control, we derive a closed-form expression for the loss function of our algorithm. In continuous control, this loss function is intractable and we therefore derive an approximation with a hyperparameter whose value regulates pessimism in Q-Learning. We present a default value for our pessimism hyperparameter that enables DoubleGum to outperform DDPG, TD3, SAC, XQL, quantile regression, and Mixture-of-Gaussian Critics in aggregate over 33 tasks from DeepMind Control, MuJoCo, MetaWorld, and Box2D and show that tuning this hyperparameter may further improve sample efficiency.

Autonomous machines must self-maintain proper functionality to ensure the safety of humans and themselves. This pertains particularly to its cameras as predominant sensors to perceive the environment and support actions. A fundamental camera problem addressed in this study is noise. Solutions often focus on denoising images a posteriori, that is, fighting symptoms rather than root causes. However, tackling root causes requires identifying the noise sources, considering the limitations of mobile platforms. This work investigates a real-time, memory-efficient and reliable noise source estimator that combines data- and physically-based models. To this end, a DNN that examines an image with camera metadata for major camera noise sources is built and trained. In addition, it quantifies unexpected factors that impact image noise or metadata. This study investigates seven different estimators on six datasets that include synthetic noise, real-world noise from two camera systems, and real field campaigns. For these, only the model with most metadata is capable to accurately and robustly quantify all individual noise contributions. This method outperforms total image noise estimators and can be plug-and-play deployed. It also serves as a basis to include more advanced noise sources, or as part of an automatic countermeasure feedback-loop to approach fully reliable machines.

Farmers must continuously improve their livestock production systems to remain competitive in the growing dairy market. Precision livestock farming technologies provide individualized monitoring of animals on commercial farms, optimizing livestock production. Continuous acoustic monitoring is a widely accepted sensing technique used to estimate the daily rumination and grazing time budget of free-ranging cattle. However, typical environmental and natural noises on pastures noticeably affect the performance limiting the practical application of current acoustic methods. In this study, we present the operating principle and generalization capability of an acoustic method called Noise-Robust Foraging Activity Recognizer (NRFAR). The proposed method determines foraging activity bouts by analyzing fixed-length segments of identified jaw movement events produced during grazing and rumination. The additive noise robustness of the NRFAR was evaluated for several signal-to-noise ratios using stationary Gaussian white noise and four different nonstationary natural noise sources. In noiseless conditions, NRFAR reached an average balanced accuracy of 86.4%, outperforming two previous acoustic methods by more than 7.5%. Furthermore, NRFAR performed better than previous acoustic methods in 77 of 80 evaluated noisy scenarios (53 cases with p<0.05). NRFAR has been shown to be effective in harsh free-ranging environments and could be used as a reliable solution to improve pasture management and monitor the health and welfare of dairy cows. The instrumentation and computational algorithms presented in this publication are protected by a pending patent application: AR P20220100910. Web demo available at: https://sinc.unl.edu.ar/web-demo/nrfar

In the context of robotics, accurate ground truth positioning is essential for the development of Simultaneous Localization and Mapping (SLAM) and control algorithms. Robotic Total Stations (RTSs) provide accurate and precise reference positions in different types of outdoor environments, especially when compared to the limited accuracy of Global Navigation Satellite System (GNSS) in cluttered areas. Three RTSs give the possibility to obtain the six-Degrees Of Freedom (DOF) reference pose of a robotic platform. However, the uncertainty of every pose is rarely computed for trajectory evaluation. As evaluation algorithms are getting increasingly precise, it becomes crucial to take into account this uncertainty. We propose a method to compute this six-DOF uncertainty from the fusion of three RTSs based on Monte Carlo (MC) methods. This solution relies on point-to-point minimization to propagate the noise of RTSs on the pose of the robotic platform. Five main noise sources are identified to model this uncertainty: noise inherent to the instrument, tilt noise, atmospheric factors, time synchronization noise, and extrinsic calibration noise. Based on extensive experimental work, we compare the impact of each noise source on the prism uncertainty and the final estimated pose. Tested on more than 50 km of trajectories, our comparison highlighted the importance of the calibration noise and the measurement distance, which should be ideally under 75 m. Moreover, it has been noted that the uncertainty on the pose of the robot is not prominently affected by one particular noise source, compared to the others.