Robust Machine Learning and the Application to Lane Change Decision Making Prediction PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download Robust Machine Learning and the Application to Lane Change Decision Making Prediction PDF full book. Access full book title Robust Machine Learning and the Application to Lane Change Decision Making Prediction by Hua Huang. Download full books in PDF and EPUB format.
Author: Hua Huang
Publisher:
ISBN:
Category : Mathematics
Languages : en
Pages : 0
Get Book Here
Book Description
In the foreseeable future, autonomous vehicles will have to drive alongside human drivers. In the absence of vehicle-to-vehicle communication, they will have to be able to predict the other road users' intentions. Equally importantly, they will also need to behave like a typical human driver such that other road users can infer their actions. It is critical to be able to learn a human driver's mental model and integrate it into the planning and control algorithm. In this dissertation, we first present a robust method to predict lane changes as cooperative or adversarial. For that, we first introduce a method to annotate lane changes as cooperative and adversarial based on the entire lane change trajectory. We then propose to train a specially designed neural network to predict the lane change label before the lane change has occurred and quantify the prediction uncertainty. The model will make lane change decisions following human drivers' driving habits and preferences, id est, it will only change lanes when the surrounding traffic is considered to be appropriate for the majority of human drivers. It will also recognize unseen novel samples and output low prediction confidence correspondingly, to alert the driver to take control or take conservative actions in such cases.
Author: Hua Huang
Publisher:
ISBN:
Category : Mathematics
Languages : en
Pages : 0
Get Book Here
Book Description
In the foreseeable future, autonomous vehicles will have to drive alongside human drivers. In the absence of vehicle-to-vehicle communication, they will have to be able to predict the other road users' intentions. Equally importantly, they will also need to behave like a typical human driver such that other road users can infer their actions. It is critical to be able to learn a human driver's mental model and integrate it into the planning and control algorithm. In this dissertation, we first present a robust method to predict lane changes as cooperative or adversarial. For that, we first introduce a method to annotate lane changes as cooperative and adversarial based on the entire lane change trajectory. We then propose to train a specially designed neural network to predict the lane change label before the lane change has occurred and quantify the prediction uncertainty. The model will make lane change decisions following human drivers' driving habits and preferences, id est, it will only change lanes when the surrounding traffic is considered to be appropriate for the majority of human drivers. It will also recognize unseen novel samples and output low prediction confidence correspondingly, to alert the driver to take control or take conservative actions in such cases.
Author: Jorge Villagra
Publisher: Elsevier
ISBN: 0323985491
Category : Technology & Engineering
Languages : en
Pages : 426
Get Book Here
Book Description
Decision-Making Techniques for Autonomous Vehicles provides a general overview of control and decision-making tools that could be used in autonomous vehicles. Motion prediction and planning tools are presented, along with the use of machine learning and adaptability to improve performance of algorithms in real scenarios. The book then examines how driver monitoring and behavior analysis are used produce comprehensive and predictable reactions in automated vehicles. The book ultimately covers regulatory and ethical issues to consider for implementing correct and robust decision-making. This book is for researchers as well as Masters and PhD students working with autonomous vehicles and decision algorithms. Provides a complete overview of decision-making and control techniques for autonomous vehicles Includes technical, physical, and mathematical explanations to provide knowledge for implementation of tools Features machine learning to improve performance of decision-making algorithms Shows how regulations and ethics influence the development and implementation of these algorithms in real scenarios
Author: Hongge Chen (Ph. D.)
Publisher:
ISBN:
Category :
Languages : en
Pages : 172
Get Book Here
Book Description
Recent studies have demonstrated that machine learning models are vulnerable to adversarial perturbations – a small and human-imperceptible input perturbation can easily change the model output completely. This has created serious security threats to many real applications, so it becomes important to formally verify the robustness of machine learning models. This thesis studies the robustness of deep neural networks as well as tree-based models, and considers the applications of robust machine learning models in deep reinforcement learning. We first develop a novel algorithm to learn robust trees. Our method aims to optimize the performance under the worst case perturbation of input features, which leads to a max-min saddle point problem when splitting nodes in trees. We propose efficient tree building algorithms by approximating the inner minimizer in this saddle point problem, and present efficient implementations for classical information gain based trees as well as state-of-the-art tree boosting models such as XGBoost. Experiments show that our method improve the model robustness significantly. We also propose an efficient method to verify the robustness of tree ensembles. We cast the tree ensembles verification problem as a max-clique problem on a multipartite graph. We develop an efficient multi-level verification algorithm that can give tight lower bounds on robustness of decision tree ensembles, while allowing iterative improvement and termination at any-time. On random forest or gradient boosted decision trees models trained on various datasets, our algorithm is up to hundreds of times faster than the previous approach that requires solving a mixed integer linear programming, and is able to give tight robustness verification bounds on large ensembles with hundreds of deep trees. For neural networks, we contribute a number of empirical studies on the practicality and the hardness of adversarial training. We show that even with adversarial defense, a model’s robustness on a test example has a strong correlation with the distance between that example and the manifold of training data embedded by the network. Test examples that are relatively far away from this manifold are more likely to be vulnerable to adversarial attacks. Consequentially, we demonstrate that an adversarial training based defense is vulnerable to a new class of attacks, the “blind-spot attack,” where the input examples reside in low density regions (“blind-spots”) of the empirical distribution of training data but are still on the valid ground-truth data manifold. Finally, we apply neural network robust training methods to deep reinforcement learning (DRL) to train agents that are robust against perturbations on state observations. We propose the state-adversarial Markov decision process (SA-MDP) to study the fundamental properties of this problem, and propose a theoretically principled regularization which can be applied to different DRL algorithms, including deep Q networks (DQN) and proximal policy optimization (PPO). We significantly improve the robustness of agents under strong white box adversarial attacks, including new attacks of our own.
Author: Maggie Makar (Computer scientist)
Publisher:
ISBN:
Category :
Languages : en
Pages : 164
Get Book Here
Book Description
We explore relationships between machine learning (ML) and causal inference. We focus on improvements in each by borrowing ideas from one another. ML has been successfully applied to many problems, but the lack of strong theoretical guarantees has led to many unexpected failures. Models that perform well on the training distribution tend to break down when applied to different distributions; small perturbations can "fool" the trained model and drastically change its predictions; arbitrary choices in the training algorithm lead to vastly different models; and so forth. On the other hand, while there has been tremendous progress in developing causal inference methods with strong theoretical guarantees, existing methods typically do not apply in practice since they assume an abundance of data. Working at the intersection of ML and causal inference, we directly address the lack of robustness in ML, and improve the statistical efficiency of causal inference techniques. The motivation behind the work presented in this thesis is to improve methods for building predictive, and causal models that are used to guide decision making. Throughout, we focus mostly on decision making in the healthcare context. On the ML for causality side, we use ML tools and analysis techniques to develop statistically efficient causal models that can guide clinicians when choosing between two treatments. On the causality for ML side, we study how knowledge of the causal mechanisms that generate observed data can be used to efficiently regularize predictive models without introducing biases. In a clinical context, we show how causal knowledge can be used to build robust, and accurate models to predict the spread of contagious infections. In a non-clinical setting, we study how to use causal knowledge to train models that are robust to distribution shifts in the context of image classification
Author: Igor Farkaš
Publisher: Springer Nature
ISBN: 3030616096
Category : Computers
Languages : en
Pages : 891
Get Book Here
Book Description
The proceedings set LNCS 12396 and 12397 constitute the proceedings of the 29th International Conference on Artificial Neural Networks, ICANN 2020, held in Bratislava, Slovakia, in September 2020.* The total of 139 full papers presented in these proceedings was carefully reviewed and selected from 249 submissions. They were organized in 2 volumes focusing on topics such as adversarial machine learning, bioinformatics and biosignal analysis, cognitive models, neural network theory and information theoretic learning, and robotics and neural models of perception and action. *The conference was postponed to 2021 due to the COVID-19 pandemic.
Author: Boris Kovalerchuk
Publisher: Springer Science & Business Media
ISBN: 0306470187
Category : Computers
Languages : en
Pages : 323
Get Book Here
Book Description
Data Mining in Finance presents a comprehensive overview of major algorithmic approaches to predictive data mining, including statistical, neural networks, ruled-based, decision-tree, and fuzzy-logic methods, and then examines the suitability of these approaches to financial data mining. The book focuses specifically on relational data mining (RDM), which is a learning method able to learn more expressive rules than other symbolic approaches. RDM is thus better suited for financial mining, because it is able to make greater use of underlying domain knowledge. Relational data mining also has a better ability to explain the discovered rules - an ability critical for avoiding spurious patterns which inevitably arise when the number of variables examined is very large. The earlier algorithms for relational data mining, also known as inductive logic programming (ILP), suffer from a relative computational inefficiency and have rather limited tools for processing numerical data. Data Mining in Finance introduces a new approach, combining relational data mining with the analysis of statistical significance of discovered rules. This reduces the search space and speeds up the algorithms. The book also presents interactive and fuzzy-logic tools for `mining' the knowledge from the experts, further reducing the search space. Data Mining in Finance contains a number of practical examples of forecasting S&P 500, exchange rates, stock directions, and rating stocks for portfolio, allowing interested readers to start building their own models. This book is an excellent reference for researchers and professionals in the fields of artificial intelligence, machine learning, data mining, knowledge discovery, and applied mathematics.
Author: Yinhai Wang
Publisher: Elsevier
ISBN: 0323996809
Category : Business & Economics
Languages : en
Pages : 254
Get Book Here
Book Description
Transportation is a combination of systems that presents a variety of challenges often too intricate to be addressed by conventional parametric methods. Increasing data availability and recent advancements in machine learning provide new methods to tackle challenging transportation problems. This textbookis designed for college or graduate-level students in transportation or closely related fields to study and understand fundamentals in machine learning. Readers will learn how to develop and apply various types of machine learning models to transportation-related problems. Example applications include traffic sensing, data-quality control, traffic prediction, transportation asset management, traffic-system control and operations, and traffic-safety analysis. Introduces fundamental machine learning theories and methodologies Presents state-of-the-art machine learning methodologies and their incorporation into transportationdomain knowledge Includes case studies or examples in each chapter that illustrate the application of methodologies andtechniques for solving transportation problems Provides practice questions following each chapter to enhance understanding and learning Includes class projects to practice coding and the use of the methods
Author: Frank Allgöwer
Publisher: Birkhäuser
ISBN: 3034884079
Category : Mathematics
Languages : en
Pages : 463
Get Book Here
Book Description
During the past decade model predictive control (MPC), also referred to as receding horizon control or moving horizon control, has become the preferred control strategy for quite a number of industrial processes. There have been many significant advances in this area over the past years, one of the most important ones being its extension to nonlinear systems. This book gives an up-to-date assessment of the current state of the art in the new field of nonlinear model predictive control (NMPC). The main topic areas that appear to be of central importance for NMPC are covered, namely receding horizon control theory, modeling for NMPC, computational aspects of on-line optimization and application issues. The book consists of selected papers presented at the International Symposium on Nonlinear Model Predictive Control – Assessment and Future Directions, which took place from June 3 to 5, 1998, in Ascona, Switzerland. The book is geared towards researchers and practitioners in the area of control engineering and control theory. It is also suited for postgraduate students as the book contains several overview articles that give a tutorial introduction into the various aspects of nonlinear model predictive control, including systems theory, computations, modeling and applications.
Author: M. Arif Wani
Publisher: Springer
ISBN: 9789811567582
Category : Technology & Engineering
Languages : en
Pages : 300
Get Book Here
Book Description
This book presents selected papers from the 18th IEEE International Conference on Machine Learning and Applications (IEEE ICMLA 2019). It focuses on deep learning networks and their application in domains such as healthcare, security and threat detection, fault diagnosis and accident analysis, and robotic control in industrial environments, and highlights novel ways of using deep neural networks to solve real-world problems. Also offering insights into deep learning architectures and algorithms, it is an essential reference guide for academic researchers, professionals, software engineers in industry, and innovative product developers.
Author: Liang Tong (Computer scientist)
Publisher:
ISBN:
Category : Machine learning
Languages : en
Pages : 0
Get Book Here
Book Description
Machine learning (ML) has come to be widely used in a broad array of settings, including important security applications such as network intrusion, fraud, and malware detection, as well as other high-stakes settings, such as autonomous driving. A general approach is to extract a set of features, or numerical attributes, of entities in question, collect a training data set of labeled examples (for example, indicating which instances are malicious and which are benign), learn a model which labels previously unseen instances presented in terms of their extracted features, and then investigate alerts raised by instances predicted as malicious. Despite the striking success of ML in security applications, security issues emerge from the full pipeline of ML-based detection systems. First, ML models are often susceptible to adversarial examples, in which an adversary makes changes to the input (such as malware) to avoid being detected. Second, using detection systems in practice is dealing with an overwhelming number of alerts that are triggered by normal behavior (the so-called false positives), obscuring alerts resulting from actual malicious activities. Third, adversaries can target a broad array of ML-based detection systems to maximize impact, which is often ignored by individual ML system designers.In this thesis, I focus on studying the security problems of deploying robust machine learning systems in adversarial settings. To conduct systematic research on this topic, my study is based on four components. First, I study the problem of systematizing adversarial evaluation. Concretely, I propose a fine-grained robustness evaluation framework for face recognition systems. Second, I investigate robust machine learning against decision-time attacks. Specifically, I propose a framework for validating models of ML evasion attacks, and evaluate the efficacy of conventional robust machine learning models against realizable attacks in PDF malware detection. My work shows that the key to robustness is the conserved features, and I propose a systematic algorithm to identify these. Additionally, I study robustness against non-salient adversarial examples in image classification and propose cognitive modeling of suspiciousness of adversarial examples. Third, I study the robust alert prioritization problem---often a necessary step in the detection pipeline. I propose a novel approach for computing a policy for prioritizing alerts using adversarial reinforcement learning. Last, I investigate robust decentralized learning, and I develop a game-theoretic model for robust linear regression involving multiple learners and a single adversary.
