Publications
2025
Random-Set Neural Networks
Shireen Kudukkil Manchingal et al. Accepted at the 13th International Conference on Learning Representations (ICLR 2025). Jan 22, 2025. [PDF]
Abstract
Machine learning is increasingly deployed in safety-critical domains where erroneous predictions may lead to potentially catastrophic consequences, highlighting the need for learning systems to be aware of how confident they are in their own predictions: in other words, 'to know when they do not know’. In this paper, we propose a novel Random-Set Neural Network (RS-NN) approach to classification which predicts belief functions (rather than classical probability vectors) over the class list using the mathematics of random sets, i.e., distributions over the collection of sets of classes. RS-NN encodes the 'epistemic' uncertainty induced by training sets that are insufficiently representative or limited in size via the size of the convex set of probability vectors associated with a predicted belief function. Our approach outperforms state-of-the-art Bayesian and Ensemble methods in terms of accuracy, uncertainty estimation and out-of-distribution (OoD) detection on multiple benchmarks (CIFAR-10 vs SVHN/Intel-Image, MNIST vs FMNIST/KMNIST, ImageNet vs ImageNet-O). RS-NN also scales up effectively to large-scale architectures (e.g. WideResNet-28-10, VGG16, Inception V3, EfficientNetB2 and ViT-Base-16), exhibits remarkable robustness to adversarial attacks and can provide statistical guarantees in a conformal learning setting.
Credal Wrapper of Model Averaging for Uncertainty Estimation in Classification
Kaizheng Wang, et. al. Accepted at the 13th International Conference on Learning Representations (ICLR 2025). Jan 22, 2025.. [PDF]
Abstract
This paper presents an innovative approach, called credal wrapper, to formulating a credal set representation of model averaging for Bayesian neural networks (BNNs) and deep ensembles (DEs), capable of improving uncertainty estimation in classification tasks. Given a finite collection of single predictive distributions derived from BNNs or DEs, the proposed credal wrapper approach extracts an upper and a lower probability bound per class, acknowledging the epistemic uncertainty due to the availability of a limited amount of distributions. Such probability intervals over classes can be mapped on a convex set of probabilities (a credal set) from which, in turn, a unique prediction can be obtained using a transformation called intersection probability transformation. In this article, we conduct extensive experiments on several out-of-distribution (OOD) detection benchmarks, encompassing various dataset pairs (CIFAR10/100 vs SVHN/Tiny-ImageNet, CIFAR10 vs CIFAR10-C, CIFAR100 vs CIFAR100-C and ImageNet vs ImageNet-O) and using different network architectures (such as VGG16, ResNet-18/50, EfficientNet B2, and ViT Base). Compared to the BNN and DE baselines, the proposed credal wrapper method exhibits superior performance in uncertainty estimation and achieves a lower expected calibration error on corrupted data.
A Unified Evaluation Framework for Epistemic Predictions
Shireen Kudukkil Manchingal, et. al. Accepted at 28th International Conference on Artificial Intelligence and Statistics (AISTATS 2025). Jan 21, 2025. [PDF]
Abstract
Predictions of uncertainty-aware models are diverse, ranging from single point estimates (often averaged over prediction samples) to predictive distributions, to set-valued or credal-set representations. We propose a novel unified evaluation framework for uncertainty-aware classifiers, applicable to a wide range of model classes, which allows users to tailor the trade-off between accuracy and precision of predictions via a suitably designed performance metric. This makes possible the selection of the most suitable model for a particular real-world application as a function of the desired trade-off. Our experiments, concerning Bayesian, ensemble, evidential, deterministic, credal and belief function classifiers on the CIFAR-10, MNIST and CIFAR-100 datasets, show that the metric behaves as desired.
CreINNs: Credal-Set Interval Neural Networks for Uncertainty Estimation in Classification Tasks
Kaizheng Wang, et. al. Neural Networks. Jan 19, 2025. [PDF]
Abstract
Effective uncertainty estimation is becoming increasingly attractive for enhancing the reliability of neural networks. This work presents a novel approach, termed Credal-Set Interval Neural Networks (CreINNs), for classification. CreINNs retain the fundamental structure of traditional Interval Neural Networks, capturing weight uncertainty through deterministic intervals. CreINNs are designed to predict an upper and a lower probability bound for each class, rather than a single probability value. The probability intervals can define a credal set, facilitating estimating different types of uncertainties associated with predictions. Experiments on standard multiclass and binary classification tasks demonstrate that the proposed CreINNs can achieve superior or comparable quality of uncertainty estimation compared to variational Bayesian Neural Networks (BNNs) and Deep Ensembles. Furthermore, CreINNs significantly reduce the computational complexity of variational BNNs during inference. Moreover, the effective uncertainty quantification of CreINNs is also verified when the input data are intervals.
2024
Credal Learning Theory
Michele Caprio, et. al. Accepted at 38th Conference on Neural Information Processing Systems (NeurIPS 2024). Sep 26, 2024. [PDF]
Abstract
Statistical learning theory is the foundation of machine learning, providing theoretical bounds for the risk of models learnt from a (single) training set, assumed to issue from an unknown probability distribution. In actual deployment, however, the data distribution may (and often does) vary, causing domain adaptation/generalization issues. In this paper we lay the foundations for a ‘credal’ theory of learning, using convex sets of probabilities (credal sets) to model the variability in the data-generating distribution. Such credal sets, we argue, may be inferred from a finite sample of training sets. Bounds are derived for the case of finite hypotheses spaces (both assuming realizability or not) as well as infinite model spaces, which directly generalize classical results.
Credal Deep Ensembles for Uncertainty Quantification
Kaizheng Wang, et. al. Accepted at 38th Conference on Neural Information Processing Systems (NeurIPS 2024). Sep 26, 2024. [PDF]
Abstract
This paper introduces an innovative approach to classification called Credal Deep Ensembles (CreDEs), namely, ensembles of novel Credal-Set Neural Networks (CreNets). CreNets are trained to predict a lower and an upper probability bound for each class, which, in turn, determine a convex set of probabilities (credal set) on the class set. The training employs a loss inspired by distributionally robust optimization which simulates the potential divergence of the test distribution from the training distribution, in such a way that the width of the predicted probability interval reflects the 'epistemic' uncertainty about the future data distribution. Ensembles can be constructed by training multiple CreNets, each associated with a different random seed, and averaging the outputted intervals.Extensive experiments are conducted on various out-of-distributions (OOD) detection benchmarks (CIFAR10/100 vs SVHN/Tiny-ImageNet, CIFAR10 vs CIFAR10-C, ImageNet vs ImageNet-O) and using different network architectures (ResNet50, VGG16, and ViT Base). Compared to Deep Ensembles baselines, CreDEs demonstrate higher test accuracy, lower expected calibration error, and significantly improved epistemic uncertainty estimation.
Bayesian Model-Free Deep Reinforcement Learning
Pascal R. van der Vaart. Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (Doctoral Consortium). September. 2024. [PDF]
Abstract
Exploration in reinforcement learning remains a difficult challenge. In order to drive exploration, ensembles with randomized prior functions have recently been popularized to quantify uncertainty in the value model. However these ensembles have no theoretical reason to resemble the actual Bayesian posterior, which is known to provide strong performance in theory under certain conditions. In this thesis work, we view training ensembles from the perspective of Sequential Monte Carlo, a Monte Carlo method that approximates a sequence of distributions with a set of particles, and propose an algorithm that exploits both the practical flexibility of ensembles and theory of the Bayesian paradigm. We incorporate this method into a standard DQN agent and experimentally show qualitatively good uncertainty quantification and improved exploration capabilities over a regular ensemble. In the future, we will investigate the impact of likelihood and prior choices in Bayesian model-free reinforcement learning methods.
Uncertainty Measures: A Critical Survey
Fabio Cuzzolin. Information Fusion. 2024. [PDF]
Abstract
Classical probability is not the only mathematical theory of uncertainty, or the most general. Many authors have argued that probability theory is ill-equipped to model the ‘epistemic’, reducible uncertainty about the process generating the data. To address this, many alternative theories of uncertainty have been formulated. In this paper, we highlight how uncertainty theories can be seen as forming clusters characterised by a shared rationale, are connected to each other in an intricate but interesting way, and can be ranked according to their degree of generality. Our objective is to propose a structured, critical summary of the research landscape in uncertainty theory, and discuss its potential for wider adoption in artificial intelligence.
Generalisation of Total Uncertainty in AI: A Theoretical Study
Keivan Shariatmadar. Preprint. Aug 1, 2024. [PDF]
Abstract
AI has been dealing with uncertainty to have highly accurate results. This becomes even worse with reasonably small data sets or a variation in the data sets. This has far-reaching effects on decision-making, forecasting and learning mechanisms. This study seeks to unpack the nature of uncertainty that exists within AI by drawing ideas from established works, the latest developments and practical applications and provide a novel total uncertainty definition in AI. From inception theories up to current methodologies, this paper provides an integrated view of dealing with better total uncertainty as well as complexities of uncertainty in AI that help us understand its meaning and value across different domains.
Robust Losses for Decision-Focused Learning
Noah Schutte, Krzysztof Postek, Neil Yorke-Smith. International Joint Conference on Artificial Intelligence (IJCAI) 2024. [PDF]
Abstract
Optimization models used to make discrete decisions often contain uncertain parameters that are context-dependent and estimated through prediction. To account for the quality of the decision made based on the prediction, decision-focused learning (end-to-end predict-then-optimize) aims at training the predictive model to minimize regret, i.e., the loss incurred by making a suboptimal decision. Despite the challenge of the gradient of this loss w.r.t. the predictive model parameters being zero almost everywhere for optimization problems with a linear objective, effective gradient-based learning approaches have been proposed to minimize the expected loss, using the empirical loss as a surrogate. However, empirical regret can be an ineffective surrogate because empirical optimal decisions can vary substantially from expected optimal decisions. To understand the impact of this deficiency, we evaluate the effect of aleatoric and epistemic uncertainty on the accuracy of empirical regret as a surrogate. Next, we propose three novel loss functions that approximate expected regret more robustly. Experimental results show that training two state-of-the-art decision-focused learning approaches using robust regret losses improves test-sample empirical regret in general while keeping computational time equivalent relative to the number of training epochs.
Science of AI – Total Uncertainty in AI
Keivan Shariatmadar. Poster. Presented at the Leuven.AI scientific workshop 2024, KU Leuven. [PDF]
Abstract
AI has been addressing uncertainty to achieve highly accurate results. This challenge becomes more pronounced with smaller data sets or variations within them, significantly impacting decision-making, forecasting, and learning mechanisms. This study explores the nature of uncertainty in AI by integrating ideas from established theories, recent developments, and practical applications, ultimately proposing a new comprehensive definition of total uncertainty in AI. From foundational theories to current methodologies, this work offers a comprehensive perspective on managing total uncertainty and its complexities in AI, helping us understand its significance and value across various domains.
A Generalisation of the Bellman Equation in Epistemic Reinforcement Learning
Keivan Shariatmadar, et. al. Poster. Presented at the Workshop on Symbolic XAI, TU Eindhoven, 2024. [PDF]
Abstract
In reinforcement learning, when a state value is unknown, we use probabilistic/stochastic MDP. Lots of work has been done regarding this problem. However, if the probability is not unique and changing, i.e., the uncertainty is epistemic, or the agent still needs to meet the state, we have less information. In this case, a novel idea is to use an epistemic uncertainty model and solve the MDP via an approach by the generalisation of the Bellman equation. In this poster, we will present this idea and show numerical results on a simple toy example.
Reinforcement Learning in Manufacturing: Investigating the Benefits of Uncertainty Awareness
Jacob Golub, et. al. Master Thesis, KU Leuven, 2024.
Abstract
This master thesis is concerned with using reinforcement learning to solve a scheduling problem in a manufacturing setting. The particular focus is on harnessing uncertainty awareness to improve the generalisability of learnt policies such that autonomous agents can navigate unfamiliar environments as well as the ones that they are trained on. The hypothesis is that by considering distributions over state-action values, the value of state features can be more easily estimated and recognised in inexperienced states. By conducting simulations of specific test scenarios, this research adds to the literature by benchmarking uncertainty aware learning approaches against deterministic ones with regard to handling unfamiliar circumstances. We show that uncertainty awareness can improve generalisability, which could be employed in manufacturing to help autonomous agents more efficiently adapt to uncertain or continuously changing factory environments.
Optimization under Epistemic Uncertainty using Prediction
Noah Schutte. International Joint Conference on Artificial Intelligence (IJCAI) 2024. Doctoral Consortium Track. [PDF]
Abstract
Due to the complexity of randomness, optimization problems are often modeled to be deterministic to be solvable. Specifically epistemic uncertainty, i.e., uncertainty that is caused due to a lack of knowledge, is not easy to model, let alone easy to subsequently solve. Despite this, taking uncertainty into account is often required for optimization models to produce robust decisions that perform well in practice. We analyze effective existing frameworks, aiming to improve robustness without increasing complexity. Specifically we focus on robustness in decision-focused learning, which is a framework aimed at making context-based predictions for an optimization problem’s uncertain parameters that minimize decision error.
Improving Metaheuristic Efficiency for Stochastic Optimization by Sequential Predictive Sampling
Noah Schutte, et al. International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research (CPAIOR 2024). May 25, 2024. [PDF]
Abstract
Metaheuristics are known to be effective in finding good solutions in combinatorial optimization, but solving stochastic problems is costly due to the need for evaluation of multiple scenarios. We propose a general method to reduce the number of scenario evaluations per solution and thus improve metaheuristic efficiency. We use a sequential sampling procedure exploiting estimates of the solutions’ expected objective values. These values are obtained with a predictive model, which is founded on an estimated discrete probability distribution linearly related to all solutions’ objective distributions; the probability distribution is continuously refined based on incoming solution evaluation. The proposed method is tested using simulated annealing, but in general applicable to single solution metaheuristics. The method’s performance is compared to descriptive sampling and an adaptation of a sequential sampling method assuming noisy evaluations. Experimental results on three problems indicate the proposed method is robust overall, and performs better on average than the baselines on two of the problems.
Diverse Projection Ensembles for Distributional Reinforcement Learning
Moritz A. Zanger, et al. The Twelfth International Conference on Learning Representations (ICLR 2024). May 07, 2024. [PDF]
Abstract
In contrast to classical reinforcement learning (RL), distributional RL algorithms aim to learn the distribution of returns rather than their expected value. Since the nature of the return distribution is generally unknown a priori or arbitrarily complex, a common approach finds approximations within a set of representable, parametric distributions. Typically, this involves a projection of the unconstrained distribution onto the set of simplified distributions. We argue that this projection step entails a strong inductive bias when coupled with neural networks and gradient descent, thereby profoundly impacting the generalization behavior of learned models. In order to facilitate reliable uncertainty estimation through diversity, we study the combination of several different projections and representations in a distributional ensemble. We establish theoretical properties of such projection ensembles and derive an algorithm that uses ensemble disagreement, measured by the average 1-Wasserstein distance, as a bonus for deep exploration. We evaluate our algorithm on the behavior suite benchmark and VizDoom and find that diverse projection ensembles lead to significant performance improvements over existing methods on a variety of tasks with the most pronounced gains in directed exploration problems.
Bayesian Ensembles for Exploration in Deep Q-Learning
Pascal Van der Vaart, Neil Yorke-Smith, and Matthijs TJ Spaan. The Sixteenth Workshop on Adaptive and Learning Agents. May 06-07 2024. [PDF]
Abstract
Exploration in reinforcement learning remains a difficult challenge. In order to drive exploration, ensembles with randomized prior functions have recently been popularized to quantify uncertainty in the value model. However these ensembles have no theoretical motivation why they should resemble the actual posterior. In this work, we view training ensembles from the perspective of Sequential Monte Carlo, a Monte Carlo method that approximates a sequence of distributions with a set of particles, and propose an algorithm that exploits both the practical flexibility of ensembles and theory of the Bayesian paradigm. We incorporate this method into a standard DQN agent and experimentally show qualitatively good uncertainty quantification and improved exploration capabilities over a regular ensemble.
Epistemic Uncertainty in Artificial Intelligence
M. Sultana and F. Cuzzolin (Editors). Lecture Notes in Artificial Intelligence, LNCS Volume 14523, April 24 2024. [PDF]
Abstract
This LNCS 14523 conference volume constitutes the proceedings of the First International Workshop, Epi UAI 2023, in Pittsburgh, PA, USA, August 2023. The 8 full papers together included in this volume were carefully reviewed and selected from 16 submissions.
On the Estimation of Image-matching Uncertainty in Visual Place Recognition
Mubariz Zaffar, Liangliang Nan and Julian F. P. Kooij. Accepted for 2024 Conference on Computer Vision and Pattern Recognition. Mar 31, 2024. [PDF]
Abstract
In Visual Place Recognition (VPR) the pose of a query image is estimated by comparing the image to a map of reference images with known reference poses. As is typical for image retrieval problems, a feature extractor maps the query and reference images to a feature space, where a nearest neighbor search is then performed. However, till recently little attention has been given to quantifying the confidence that a retrieved reference image is a correct match. Highly certain but incorrect retrieval can lead to catastrophic failure of VPR-based localization pipelines. This work compares for the first time the main approaches for estimating the image-matching uncertainty, including the traditional retrieval-based uncertainty estimation, more recent data-driven aleatoric uncertainty estimation, and the compute-intensive geometric verification. We further formulate a simple baseline method, ''SUE'', which unlike the other methods considers the freely-available poses of the reference images in the map. Our experiments reveal that a simple L2-distance between the query and reference descriptors is already a better estimate of image-matching uncertainty than current data-driven approaches. SUE outperforms the other efficient uncertainty estimation methods, and its uncertainty estimates complement the computationally expensive geometric verification approach. Future works for uncertainty estimation in VPR should consider the baselines discussed in this work.
Generalising Realisability in Statistical Learning Theory under Epistemic Uncertainty
Fabio Cuzzolin. Preprint. Feb 22, 2024. [PDF]
Abstract
The purpose of this paper is to look into how central notions in statistical learning theory, such as realisability, generalise under the assumption that train and test distribution are issued from the same credal set, i.e., a convex set of probability distributions. This can be considered as a first step towards a more general treatment of statistical learning under epistemic uncertainty.
A Hybrid Graph Network for Complex Activity Detection in Video
Salman Khan, et. al. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024. [PDF]
Abstract
Interpretation and understanding of video presents a challenging computer vision task in numerous fields- e.g. autonomous driving and sports analytics. Existing approaches to interpreting the actions taking place within a video clip are based upon Temporal Action Localisation (TAL), which typically identifies short-term actions. The emerging field of Complex Activity Detection (CompAD) extends this analysis to long-term activities, with a deeper understanding obtained by modelling the internal structure of a complex activity taking place within the video. We address the CompAD problem using a hybrid graph neural network which combinesattention applied to a graph encoding the local (short-term) dynamic scene with a temporal graph modelling the overall long-duration activity. Our approach is as follows: i) Firstly, we propose a novel feature extraction technique which, for each video snippet, generates spatiotemporal ‘tubes’ for the active elements (‘agents’) in the (local) scene by detecting individual objects, tracking them and then extracting 3D features from all the agent tubes as well as the overall scene. ii) Next, we construct a local scene graph where each node (representing either an agent tube or the scene) is connected to all other nodes. Attention is then applied to this graph to obtain an overall representation of the local dynamic scene. iii) Finally, all local scene graph representations are interconnected via a temporal graph, to estimate the complex activity class together with its start and end time. The proposed framework outperforms all previous state-of-the-art methods on all three datasets including ActivityNet-1.3, Thumos-14, and ROAD.
Interval Reduced Order Surrogate Modelling Framework for Uncertainty Quantification
Ghifari A. Faza, Keivan Shariatmadar, Hans Hallez and David Moens. AIAA Scitech 2024 Forum. Jan 4, 2024. [PDF]
Abstract
Surrogate models are widely used in the engineering community to approximate costly and large evaluation processes, such as difficult experiments or expensive simulations. This paper presents a non-intrusive framework for epistemic surrogate modelling, which is based on proper orthogonal decomposition (POD) and polynomial chaos expansion (PCE) for interval observations. In physical systems modelling, it is important to consider both aleatoric and epistemic uncertainty by constructing an uncertainty model to be included in the surrogate model. However, existing frameworks have a major limitation in that they can only handle scalar data observations and are not designed for non-deterministic observations like interval data. In many applications, the observed data can be inherently non-scalar due to various factors, such as observation uncertainties, conflicting data, or summarized data. In our proposed framework, we integrate POD for interval data with PCE for interval observations. Firstly, we employ interval POD to obtain an optimally reduced-order basis from the full-order snapshot. Then, we approximate this reduced-order basis using a non-intrusive interval PCE method. Allowing non-scalar data, such as intervals, is advantageous as it takes into account more information in the physical system modelling.
2023
Reasoning with Random Sets: An Agenda for the Future
Fabio Cuzzolin. Preprint. December 19, 2023. [PDF]
Abstract
In this paper, we discuss a potential agenda for future work in the theory of random sets and belief functions, touching upon a number of focal issues: the development of a fully-fledged theory of statistical reasoning with random sets, including the generalisation of logistic regression and of the classical laws of probability; the further development of the geometric approach to uncertainty, to include general random sets, a wider range of uncertainty measures and alternative geometric representations; the application of this new theory to high-impact areas such as climate change, machine learning and statistical learning theory.
Bayesian Deep Q-Learning via Sequential Monte Carlo
Pascal Van der Vaart, Matthijs T. J. Spaan, Neil Yorke-Smith. Proceedings of the 16th European Workshop on Reinforcement Learning, Brussels, Belgium, September 2023. [PDF]
Abstract
Exploration in reinforcement learning remains a difficult challenge. Recently, ensembles with randomized prior functions have been popularized to quantify uncertainty in the value model, in order to drive exploration with success. However these ensembles have no theoretical guarantee to resemble the actual posterior. In this work, we view training ensembles from the perspective of sequential Monte Carlo, and propose an algorithm that exploits both the practical flexibility of ensembles and theory of the Bayesian paradigm. We incorporate this method into a standard DQN agent and experimentally show improved exploration capabilities over a regular ensemble.
E-MCTS: Deep Exploration in Model-Based Reinforcement Learning by Planning with Epistemic Uncertainty
Yaniv Oren et al. 16th European Workshop on Reinforcement Learning. 2023. Aug 30, 2023. [PDF]
Abstract
One of the most well-studied and highly performing planning approaches used in Model-Based Reinforcement Learning (MBRL) is Monte-Carlo Tree Search (MCTS). Key challenges of MCTS-based MBRL methods remain dedicated deep exploration and reliability in the face of the unknown, and both challenges can be alleviated through principled epistemic uncertainty estimation in the predictions of MCTS. We present two main contributions: First, we develop methodology to propagate epistemic uncertainty in MCTS, enabling agents to estimate the epistemic uncertainty in their predictions. Second, we utilize the propagated uncertainty for a novel deep exploration algorithm by explicitly planning to explore. We incorporate our approach into variations of MCTS-based MBRL approaches with learned and provided models, and empirically show deep exploration through successful epistemic uncertainty estimation achieved by our approach. We compare to a non-planning-based deep-exploration baseline, and demonstrate that planning with epistemic MCTS significantly outperforms non-planning based exploration in the investigated setting.
Fleet Planning under Demand and Fuel Price Uncertainty using Actor–critic Reinforcement Learning
Izaak L. Geursen et al. Journal of Air Transport Management. June, 2023. [PDF]
Abstract
Current state-of-the-art airline planning models face computational limitations, restricting the operational applicability to problems of representative sizes. This is particularly the case when considering the uncertainty necessarily associated with the long-term plan of an aircraft fleet. Considering the growing interest in the application of machine learning techniques to operations research problems, this article investigates the applicability of these techniques for airline planning. Specifically, an Advantage Actor–Critic (A2C) reinforcement learning algorithm is developed for the airline fleet planning problem. The increased computational efficiency of using an A2C agent allows us to consider real-world-sized problems and account for highly-volatile uncertainty in demand and fuel price. The result is a multi-stage probabilistic fleet plan describing the evolution of the fleet according to a large set of future scenarios. The A2C algorithm is found to outperform a deterministic model and a deep Q-network algorithm. The relative performance of the A2C increases as more complexity is added to the problem. Further, the A2C algorithm can compute a multi-stage fleet planning solution within a few seconds.
ROAD-R: The Autonomous Driving Dataset with Logical Requirements
Eleonora Giunchiglia et al. Machine Learning. May 1, 2023. [PDF]
Abstract
Neural networks have proven to be very powerful at computer vision tasks. However, they often exhibit unexpected behaviors, acting against background knowledge about the problem at hand. This calls for models (i) able to learn from requirements expressing such background knowledge, and (ii) guaranteed to be compliant with the requirements themselves. Unfortunately, the development of such models is hampered by the lack of real-world datasets equipped with formally specified requirements. In this paper, we introduce the ROad event Awareness Dataset with logical Requirements (ROAD-R), the first publicly available dataset for autonomous driving with requirements expressed as logical constraints. Given ROAD-R, we show that current state-of-the-art models often violate its logical constraints, and that it is possible to exploit them to create models that (i) have a better performance, and (ii) are guaranteed to be compliant with the requirements themselves.
2022
Path Planning Problem under Non-probabilistic Uncertainty
Shariatmadar, Keivan. Preprint. Dec 1, 2022. [PDF]
Abstract
This paper considers theoretical solutions for path planning problems under non-probabilistic uncertainty used in the travel salesman problems under uncertainty. The uncertainty is on the paths between the cities as nodes in a travelling salesman problem. There is at least one path between two nodes/stations where the travelling time between the nodes is not precisely known. This could be due to environmental effects like crowdedness (rush period) in the path, the state of the charge of batteries, weather conditions, or considering the safety of the route while travelling. In this work, we consider two different advanced uncertainty models (i) probabilistic-precise uncertain model: Probability distributions and (ii) non-probabilistic--imprecise uncertain model: Intervals. We investigate what theoretical results can be obtained for two different optimality criteria: maximinity and maximality in the travelling salesman problem.
An Introduction to Optimization under Uncertainty--A Short Survey
Shariatmadar, Keivan, et al. Preprint. Dec 1, 2022. [PDF]
Abstract
Optimization equips engineers and scientists in a variety of fields with the ability to transcribe their problems into a generic formulation and receive optimal solutions with relative ease. Industries ranging from aerospace to robotics continue to benefit from advancements in optimization theory and the associated algorithmic developments. Nowadays, optimization is used in real time on autonomous systems acting in safety-critical situations, such as self-driving vehicles. It has become increasingly more important to produce robust solutions by incorporating uncertainty into optimization programs. This paper provides a short survey about the state of the art in optimization under uncertainty. The paper begins with a brief overview of the main classes of optimization without uncertainty. The rest of the paper focuses on the different methods for handling both aleatoric and epistemic uncertainty. Many of the applications discussed in this paper are within the domain of control. The goal of this survey paper is to briefly touch upon the state of the art in a variety of different methods and refer the reader to other literature for more in-depth treatments of the topics discussed here.
Berth Planning and Real‑time Disruption Recovery: A Simulation Study for A Tidal Port
Jaap‑Jan van der Steeg, et al. Flexible Services and Manufacturing Journal. Nov 5, 2022. [PDF]
Abstract
With the increasing volume of container freight transport, future port planning is crucial. Simulation models provide a means to gain insight in the efects of terminal expansions. Detailed simulations incorporate berth allocation: assigning vessels a time and location at the quay wall, where the vessel is loaded and unloaded. This article develops decision models for both ofine preliminary berth planning and for online recovery of this plan during simulation. First, we develop an optimisation-based approach that incorporates realistic aspects—cyclic vessel arrivals, tidal windows, and minimisation of vessel draught during low water periods—in order to develop a cyclic baseline berth allocation plan. The approach can proactively incorporate slack for increased robustness. Exploiting a constraint-based solver, we can obtain optimal or satisfcing solutions for a year’s operation of a large port. The resulting preliminary berth plan is used as a basis for the arrival times. However, disruptions can occur, such as vessel arrival and loading times varying from the planned. Hence, second, we develop a real-time disruption management decision model. This multi-level heuristic approach reacts to disruptions while minimising perturbation of the original berth plan. Computational experiments with a high-resolution simulator show our recovery approach fnds good solutions until a tipping point of disturbance. Results also show that when the expected occupation of a terminal is higher, strengthening robustness of the preliminary plan has increased importance. The approach described in the article is implemented for a major European inland tidal port, forming the basis of a simulation-based decision support tool for operational planning and exploring port expansion options.
Epistemic Deep Learning
Manchingal, Shireen Kudukkil and Cuzzolin, Fabio. Presented at ICML 2022 Workshop on Distribution-free Uncertainty Quantification. Jul 23, 2022. [PDF]
Abstract
The belief function approach to uncertainty quantification as proposed in the Demspter-Shafer theory of evidence is established upon the general mathematical models for set-valued observations, called random sets. Set-valued predictions are the most natural representations of uncertainty in machine learning. In this paper, we introduce a concept called epistemic deep learning based on the random-set interpretation of belief functions to model epistemic learning in deep neural networks. We propose a novel random-set convolutional neural network for classification that produces scores for sets of classes by learning set-valued ground truth representations. We evaluate different formulations of entropy and distance measures for belief functions as viable loss functions for these random-set networks. We also discuss methods for evaluating the quality of epistemic predictions and the performance of epistemic random-set neural networks. We demonstrate through experiments that the epistemic approach produces better performance results when compared to traditional approaches of estimating uncertainty.
Theory of Mind and Preference Learning at the Interface of Cognitive Science, Neuroscience, and AI: A Review
Langley, Christelle, et al. Frontiers in Artificial Intelligence. Apr 05, 2022. [PDF]
Abstract
Theory of Mind (ToM)—the ability of the human mind to attribute mental states to others—is a key component of human cognition. In order to understand other people's mental states or viewpoint and to have successful interactions with others within social and occupational environments, this form of social cognition is essential. The same capability of inferring human mental states is a prerequisite for artificial intelligence (AI) to be integrated into society, for example in healthcare and the motoring industry. Autonomous cars will need to be able to infer the mental states of human drivers and pedestrians to predict their behavior. In the literature, there has been an increasing understanding of ToM, specifically with increasing cognitive science studies in children and in individuals with Autism Spectrum Disorder. Similarly, with neuroimaging studies there is now a better understanding of the neural mechanisms that underlie ToM. In addition, new AI algorithms for inferring human mental states have been proposed with more complex applications and better generalisability. In this review, we synthesize the existing understanding of ToM in cognitive and neurosciences and the AI computational models that have been proposed. We focus on preference learning as an area of particular interest and the most recent neurocognitive and computational ToM models. We also discuss the limitations of existing models and hint at potential approaches to allow ToM models to fully express the complexity of the human mind in all its aspects, including values and preferences.
Road: The ROad event Awareness Dataset for autonomous Driving
Singh, Gurkirt, et al. IEEE Transactions on Pattern Analysis and Machine Intelligence. Feb 14 2022. [PDF]
Abstract
Humans drive in a holistic fashion which entails, in particular, understanding dynamic road events and their evolution. Injecting these capabilities in autonomous vehicles can thus take situational awareness and decision making closer to human-level performance. To this purpose, we introduce the ROad event Awareness Dataset (ROAD) for Autonomous Driving, to our knowledge the first of its kind. ROAD is designed to test an autonomous vehicles ability to detect road events, defined as triplets composed by an active agent, the action(s) it performs and the corresponding scene locations. ROAD comprises videos originally from the Oxford RobotCar Dataset annotated with bounding boxes showing the location in the image plane of each road event. We benchmark various detection tasks, proposing as a baseline a new incremental algorithm for online road event awareness termed 3D-RetinaNet. We also report the performance on the ROAD tasks of Slowfast and YOLOv5 detectors, as well as that of the winners of the ICCV2021 ROAD challenge, which highlight the challenges faced by situation awareness in autonomous driving. ROAD is designed to allow scholars to investigate exciting tasks such as complex (road) activity detection, future event anticipation and continual learning. The dataset is available at https://github.com/gurkirt/road-dataset; the baseline can be found at https://github.com/gurkirt/3D-RetinaNet.
Vision in adverse weather: Augmentation using CycleGANs with various object detectors for robust perception in autonomous racing
Teeti, Izzeddin, et al. Preprint. Jan 11 2022. [PDF]
Abstract
In an autonomous driving system, perception - identification of features and objects from the environment - is crucial. In autonomous racing, high speeds and small margins demand rapid and accurate detection systems. During the race, the weather can change abruptly, causing significant degradation in perception, resulting in ineffective manoeuvres. In order to improve detection in adverse weather, deep-learning-based models typically require extensive datasets captured in such conditions - the collection of which is a tedious, laborious, and costly process. However, recent developments in CycleGAN architectures allow the synthesis of highly realistic scenes in multiple weather conditions. To this end, we introduce an approach of using synthesised adverse condition datasets in autonomous racing (generated using CycleGAN) to improve the performance of four out of five state-of-the-art detectors by an average of 42.7 and 4.4 mAP percentage points in the presence of night-time conditions and droplets, respectively. Furthermore, we present a comparative analysis of five object detectors - identifying the optimal pairing of detector and training data for use during autonomous racing in challenging conditions.
The Intersection Probability: Betting with Probability Intervals
Cuzzolin, Fabio. Preprint. Jan 5 2022. [PDF]
Abstract
Probability intervals are an attractive tool for reasoning under uncertainty. Unlike belief functions, though, they lack a natural probability transformation to be used for decision making in a utility theory framework. In this paper we propose the use of the intersection probability, a transform derived originally for belief functions in the framework of the geometric approach to uncertainty, as the most natural such transformation. We recall its rationale and definition, compare it with other candidate representives of systems of probability intervals, discuss its credal rationale as focus of a pair of simplices in the probability simplex, and outline a possible decision making framework for probability intervals, analogous to the Transferable Belief Model for belief functions.
2021
YOLO-Z: Improving Small Object detection in YOLOv5 for Autonomous Vehicles
Benjumea, Aduen, et al. ICCV Workshop: The ROAD challenge: Event Detection for Situation Awareness in Autonomous Driving. Dec 23 2021. [PDF]
Abstract
As autonomous vehicles and autonomous racing rise in popularity, so does the need for faster and more accurate detectors. While our naked eyes are able to extract contextual information almost instantly, even from far away, image resolution and computational resources limitations make detecting smaller objects (that is, objects that occupy a small pixel area in the input image) a genuinely challenging task for machines and a wide-open research field. This study explores how the popular YOLOv5 object detector can be modified to improve its performance in detecting smaller objects, with a particular application in autonomous racing. To achieve this, we investigate how replacing certain structural elements of the model (as well as their connections and other parameters) can affect performance and inference time. In doing so, we propose a series of models at different scales, which we name `YOLO-Z', and which display an improvement of up to 6.9% in mAP when detecting smaller objects at 50% IOU, at the cost of just a 3ms increase in inference time compared to the original YOLOv5. Our objective is to inform future research on the potential of adjusting a popular detector such as YOLOv5 to address specific tasks and provide insights on how specific changes can impact small object detection. Such findings, applied to the broader context of autonomous vehicles, could increase the amount of contextual information available to such systems.
DeepSmoke: Deep Learning Model for Smoke Detection and Segmentation in Outdoor Environments
Khan, Salman, et al. Expert Systems with Applications. Volume 182. Nov 15 2021. [PDF]
Abstract
Fire disaster throughout the globe causes social, environmental, and economical damage, making its early detection and instant reporting essential for saving human lives and properties. Smoke detection plays a key role in early fire detection but majority of the existing methods are limited to either indoor or outdoor surveillance environments, with poor performance for hazy scenarios. In this paper, we present a Convolutional Neural Network (CNN)-based smoke detection and segmentation framework for both clear and hazy environments. Unlike existing methods, we employ an efficient CNN architecture, termed EfficientNet, for smoke detection with better accuracy. We also segment the smoke regions using DeepLabv3+, which is supported by effective encoders and decoders along with a pixel-wise classifier for optimum localization. Our smoke detection results evince a noticeable gain up to 3% in accuracy and a decrease of 0.46% in False Alarm Rate (FAR), while segmentation reports a significant increase of 2% and 1% in global accuracy and mean Intersection over Union (IoU) scores, respectively. This makes our method a best fit for smoke detection and segmentation in real-world surveillance settings.
Multi-weather City: Adverse Weather Stacking for Autonomous Driving
Mușat, Valentina, et al. 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW). Oct 11-17 2021. [PDF]
Abstract
Autonomous vehicles make use of sensors to perceive the world around them, with heavy reliance on vision based sensors such as RGB cameras. Unfortunately, since these sensors are affected by adverse weather, perception pipelines require extensive training on visual data under harsh conditions in order to improve the robustness of downstream tasks - data that is difficult and expensive to acquire. Based on GAN and CycleGAN architectures, we propose an overall (modular) architecture for constructing datasets, which allows one to add, swap out and combine components in order to generate images with diverse weather conditions. Starting from a single dataset with ground-truth, we generate 7 versions of the same data in diverse weather, and propose an extension to augment the generated conditions, thus resulting in a total of 14 adverse weather conditions, requiring a single ground truth. We test the quality of the generated conditions both in terms of perceptual quality and suitability for training downstream tasks, using real world, out-of-distribution adverse weather extracted from various datasets. We show improvements in both object detection and instance segmentation across all conditions, in many cases exceeding 10 percentage points increase in AP, and provide the materials and instructions needed to re-construct the multi-weather dataset, based upon the original Cityscapes dataset.
A Geometric Approach to Conditioning Belief Functions
Cuzzolin, Fabio. Preprint. Apr 21 2021. [PDF]
Abstract
Conditioning is crucial in applied science when inference involving time series is involved. Belief calculus is an effective way of handling such inference in the presence of epistemic uncertainty -- unfortunately, different approaches to conditioning in the belief function framework have been proposed in the past, leaving the matter somewhat unsettled. Inspired by the geometric approach to uncertainty, in this paper we propose an approach to the conditioning of belief functions based on geometrically projecting them onto the simplex associated with the conditioning event in the space of all belief functions. We show here that such a geometric approach to conditioning often produces simple results with straightforward interpretations in terms of degrees of belief. This raises the question of whether classical approaches, such as for instance Dempster's conditioning, can also be reduced to some form of distance minimisation in a suitable space. The study of families of combination rules generated by (geometric) conditioning rules appears to be the natural prosecution of the presented research.
Uncertainty Measures: The Big Picture
Cuzzolin, Fabio. Preprint. Apr 14 2021. [PDF]
Abstract
Probability theory is far from being the most general mathematical theory of uncertainty. A number of arguments point at its inability to describe second-order ('Knightian') uncertainty. In response, a wide array of theories of uncertainty have been proposed, many of them generalisations of classical probability. As we show here, such frameworks can be organised into clusters sharing a common rationale, exhibit complex links, and are characterised by different levels of generality. Our goal is a critical appraisal of the current landscape in uncertainty theory.