Alberto Sangiovanni‐Vincentelli

Min Liu, Alberto Sangiovanni-Vincentelli, Xiangyu Yue · berkeley

Deep neural networks (DNNs) are vulnerable to backdoor attack, which does not affect the network's performance on clean data but would manipulate the network behavior once a trigger pattern is added. Existing defense methods have greatly reduced attack success rate, but their prediction accuracy on clean data still lags behind a clean model by a large margin. Inspired by the stealthiness and effectiveness of backdoor attack, we propose a simple but highly effective defense framework which injects non-adversarial backdoors targeting poisoned samples. Following the general steps in backdoor attack, we detect a small set of suspected samples and then apply a poisoning strategy to them. The non-adversarial backdoor, once triggered, suppresses the attacker's backdoor on poisoned data, but has limited influence on clean data. The defense can be carried out during data preprocessing, without any modification to the standard end-to-end training pipeline. We conduct extensive experiments on multiple benchmarks with different architectures and representative attacks. Results demonstrate that our method achieves state-of-the-art defense effectiveness with by far the lowest performance drop on clean data. Considering the surprising defense ability displayed by our framework, we call for more attention to utilizing backdoor for backdoor defense. Code is available at https://github.com/minliu01/non-adversarial_backdoor.

Vasumathi Raman, Alexandre Donzé, Mehdi Maasoumy et al.

We present a mathematical programming-based method for model predictive control of cyber-physical systems subject to signal temporal logic (STL) specifications. We describe the use of STL to specify a wide range of properties of these systems, including safety, response and bounded liveness. For synthesis, we encode STL specifications as mixed integer-linear constraints on the system variables in the optimization problem at each step of a receding horizon control framework. We prove correctness of our algorithms, and present experimental results for controller synthesis for building energy and climate control.

Shromona Ghosh, Dorsa Sadigh, Pierluigi Nuzzo et al.

We address the problem of diagnosing and repairing specifications for hybrid systems formalized in signal temporal logic (STL). Our focus is on the setting of automatic synthesis of controllers in a model predictive control (MPC) framework. We build on recent approaches that reduce the controller synthesis problem to solving one or more mixed integer linear programs (MILPs), where infeasibility of a MILP usually indicates unrealizability of the controller synthesis problem. Given an infeasible STL synthesis problem, we present algorithms that provide feedback on the reasons for unrealizability, and suggestions for making it realizable. Our algorithms are sound and complete, i.e., they provide a correct diagnosis, and always terminate with a non-trivial specification that is feasible using the chosen synthesis method, when such a solution exists. We demonstrate the effectiveness of our approach on the synthesis of controllers for various cyber-physical systems, including an autonomous driving application and an aircraft electric power system.

Mirco Theile, Harald Bayerlein, Marco Caccamo et al.

Coverage path planning (CPP) is a critical problem in robotics, where the goal is to find an efficient path that covers every point in an area of interest. This work addresses the power-constrained CPP problem with recharge for battery-limited unmanned aerial vehicles (UAVs). In this problem, a notable challenge emerges from integrating recharge journeys into the overall coverage strategy, highlighting the intricate task of making strategic, long-term decisions. We propose a novel proximal policy optimization (PPO)-based deep reinforcement learning (DRL) approach with map-based observations, utilizing action masking and discount factor scheduling to optimize coverage trajectories over the entire mission horizon. We further provide the agent with a position history to handle emergent state loops caused by the recharge capability. Our approach outperforms a baseline heuristic, generalizes to different target zones and maps, with limited generalization to unseen maps. We offer valuable insights into DRL algorithm design for long-horizon problems and provide a publicly available software framework for the CPP problem.

Mirco Theile, Daniele Bernardini, Raphael Trumpp et al.

Modern cyber-physical systems are becoming increasingly complex to model, thus motivating data-driven techniques such as reinforcement learning (RL) to find appropriate control agents. However, most systems are subject to hard constraints such as safety or operational bounds. Typically, to learn to satisfy these constraints, the agent must violate them systematically, which is computationally prohibitive in most systems. Recent efforts aim to utilize feasibility models that assess whether a proposed action is feasible to avoid applying the agent's infeasible action proposals to the system. However, these efforts focus on guaranteeing constraint satisfaction rather than the agent's learning efficiency. To improve the learning process, we introduce action mapping, a novel approach that divides the learning process into two steps: first learn feasibility and subsequently, the objective by mapping actions into the sets of feasible actions. This paper focuses on the feasibility part by learning to generate all feasible actions through self-supervised querying of the feasibility model. We train the agent by formulating the problem as a distribution matching problem and deriving gradient estimators for different divergences. Through an illustrative example, a robotic path planning scenario, and a robotic grasping simulation, we demonstrate the agent's proficiency in generating actions across disconnected feasible action sets. By addressing the feasibility step, this paper makes it possible to focus future work on the objective part of action mapping, paving the way for an RL framework that is both safe and efficient.

Piergiuseppe Mallozzi, Hussein Sibai, Inigo Incer et al.

We propose a context-sensitive grammar for the systematic exploration of the design space of the topology of 3D robots, particularly unmanned aerial vehicles. It defines production rules for adding components to an incomplete design topology modeled over a 3D grid. The rules are local. The grammar is simple, yet capable of modeling most existing UAVs as well as novel ones. It can be easily generalized to other robotic platforms. It can be thought of as a building block for any design exploration and optimization algorithm.

Carlo Bosio, Matteo Guarrera, Alberto Sangiovanni-Vincentelli et al.

Large Language models (LLMs) have shown promise as generators of symbolic control policies, producing interpretable program-like representations through iterative search. However, these models are not capable of separating the functional structure of a policy from the numerical values it is parametrized by, thus making the search process slow and inefficient. We propose a hybrid approach that decouples structural synthesis from parameter optimization by introducing an additional optimization layer for local parameter search. In our method, the numerical parameters of LLM-generated programs are extracted and optimized numerically to maximize task performance. With this integration, an LLM iterates over the functional structure of programs, while a separate optimization loop is used to find a locally optimal set of parameters accompanying candidate programs. We evaluate our method on a set of control tasks, showing that it achieves higher returns and improved sample efficiency compared to purely LLM-guided search. We show that combining symbolic program synthesis with numerical optimization yields interpretable yet high-performing policies, bridging the gap between language-model-guided design and classical control tuning. Our code is available at https://sites.google.com/berkeley.edu/colmo.

Karim Elmaaroufi, Devan Shanker, Ana Cismaru et al. · berkeley, cmu

For cyber-physical systems (CPS), including robotics and autonomous vehicles, mass deployment has been hindered by fatal errors that occur when operating in rare events. To replicate rare events such as vehicle crashes, many companies have created logging systems and employed crash reconstruction experts to meticulously recreate these valuable events in simulation. However, in these methods, "what if" questions are not easily formulated and answered. We present ScenarioNL, an AI System for creating scenario programs from natural language. Specifically, we generate these programs from police crash reports. Reports normally contain uncertainty about the exact details of the incidents which we represent through a Probabilistic Programming Language (PPL), Scenic. By using Scenic, we can clearly and concisely represent uncertainty and variation over CPS behaviors, properties, and interactions. We demonstrate how commonplace prompting techniques with the best Large Language Models (LLM) are incapable of reasoning about probabilistic scenario programs and generating code for low-resource languages such as Scenic. Our system is comprised of several LLMs chained together with several kinds of prompting strategies, a compiler, and a simulator. We evaluate our system on publicly available autonomous vehicle crash reports in California from the last five years and share insights into how we generate code that is both semantically meaningful and syntactically correct.

Mirco Theile, Lukas Dirnberger, Raphael Trumpp et al.

Deep reinforcement learning (DRL) has had success across various domains, but applying it to environments with constraints remains challenging due to poor sample efficiency and slow convergence. Recent literature explored incorporating model knowledge to mitigate these problems, particularly through the use of models that assess the feasibility of proposed actions. However, integrating feasibility models efficiently into DRL pipelines in environments with continuous action spaces is non-trivial. We propose a novel DRL training strategy utilizing action mapping that leverages feasibility models to streamline the learning process. By decoupling the learning of feasible actions from policy optimization, action mapping allows DRL agents to focus on selecting the optimal action from a reduced feasible action set. We demonstrate through experiments that action mapping significantly improves training performance in constrained environments with continuous action spaces, especially with imperfect feasibility models.

Mirco Theile, Hongpeng Cao, Marco Caccamo et al.

In reinforcement learning (RL), exploiting environmental symmetries can significantly enhance efficiency, robustness, and performance. However, ensuring that the deep RL policy and value networks are respectively equivariant and invariant to exploit these symmetries is a substantial challenge. Related works try to design networks that are equivariant and invariant by construction, limiting them to a very restricted library of components, which in turn hampers the expressiveness of the networks. This paper proposes a method to construct equivariant policies and invariant value functions without specialized neural network components, which we term equivariant ensembles. We further add a regularization term for adding inductive bias during training. In a map-based path planning case study, we show how equivariant ensembles and regularization benefit sample efficiency and performance.

Edward Kim, Jay Shenoy, Sebastian Junges et al.

Simulation-based testing of autonomous vehicles (AVs) has become an essential complement to road testing to ensure safety. Consequently, substantial research has focused on searching for failure scenarios in simulation. However, a fundamental question remains: are AV failure scenarios identified in simulation meaningful in reality, i.e., are they reproducible on the real system? Due to the sim-to-real gap arising from discrepancies between simulated and real sensor data, a failure scenario identified in simulation can be either a spurious artifact of the synthetic sensor data or an actual failure that persists with real sensor data. An approach to validate simulated failure scenarios is to identify instances of the scenario in a corpus of real data, and check if the failure persists on the real data. To this end, we propose a formal definition of what it means for a labelled data item to match an abstract scenario, encoded as a scenario program using the SCENIC probabilistic programming language. Using this definition, we develop a querying algorithm which, given a scenario program and a labelled dataset, finds the subset of data matching the scenario. Experiments demonstrate that our algorithm is accurate and efficient on a variety of realistic traffic scenarios, and scales to a reasonable number of agents.

Matteo Guarrera, Baihong Jin, Tung-Wei Lin et al.

We consider the problem of detecting OoD(Out-of-Distribution) input data when using deep neural networks, and we propose a simple yet effective way to improve the robustness of several popular OoD detection methods against label shift. Our work is motivated by the observation that most existing OoD detection algorithms consider all training/test data as a whole, regardless of which class entry each input activates (inter-class differences). Through extensive experimentation, we have found that such practice leads to a detector whose performance is sensitive and vulnerable to label shift. To address this issue, we propose a class-wise thresholding scheme that can apply to most existing OoD detection algorithms and can maintain similar OoD detection performance even in the presence of label shift in the test distribution.

Xiangyu Yue, Zangwei Zheng, Colorado Reed et al.

Multi-source Domain Adaptation (MDA) aims to transfer predictive models from multiple, fully-labeled source domains to an unlabeled target domain. However, in many applications, relevant labeled source datasets may not be available, and collecting source labels can be as expensive as labeling the target data itself. In this paper, we investigate Multi-source Few-shot Domain Adaptation (MFDA): a new domain adaptation scenario with limited multi-source labels and unlabeled target data. As we show, existing methods often fail to learn discriminative features for both source and target domains in the MFDA setting. Therefore, we propose a novel framework, termed Multi-Source Few-shot Adaptation Network (MSFAN), which can be trained end-to-end in a non-adversarial manner. MSFAN operates by first using a type of prototypical, multi-domain, self-supervised learning to learn features that are not only domain-invariant but also class-discriminative. Second, MSFAN uses a small, labeled support set to enforce feature consistency and domain invariance across domains. Finally, prototypes from multiple sources are leveraged to learn better classifiers. Compared with state-of-the-art MDA methods, MSFAN improves the mean classification accuracy over different domain pairs on MFDA by 20.2%, 9.4%, and 16.2% on Office, Office-Home, and DomainNet, respectively.

Hari Prasanna Das, Ryan Tran, Japjot Singh et al.

$\textbf{Background:}$ At the onset of a pandemic, such as COVID-19, data with proper labeling/attributes corresponding to the new disease might be unavailable or sparse. Machine Learning (ML) models trained with the available data, which is limited in quantity and poor in diversity, will often be biased and inaccurate. At the same time, ML algorithms designed to fight pandemics must have good performance and be developed in a time-sensitive manner. To tackle the challenges of limited data, and label scarcity in the available data, we propose generating conditional synthetic data, to be used alongside real data for developing robust ML models. $\textbf{Methods:}$ We present a hybrid model consisting of a conditional generative flow and a classifier for conditional synthetic data generation. The classifier decouples the feature representation for the condition, which is fed to the flow to extract the local noise. We generate synthetic data by manipulating the local noise with fixed conditional feature representation. We also propose a semi-supervised approach to generate synthetic samples in the absence of labels for a majority of the available data. $\textbf{Results:}$ We performed conditional synthetic generation for chest computed tomography (CT) scans corresponding to normal, COVID-19, and pneumonia afflicted patients. We show that our method significantly outperforms existing models both on qualitative and quantitative performance, and our semi-supervised approach can efficiently synthesize conditional samples under label scarcity. As an example of downstream use of synthetic data, we show improvement in COVID-19 detection from CT scans with conditional synthetic data augmentation.

Zangwei Zheng, Xiangyu Yue, Kurt Keutzer et al.

Object detection is essential to safe autonomous or assisted driving. Previous works usually utilize RGB images or LiDAR point clouds to identify and localize multiple objects in self-driving. However, cameras tend to fail in bad driving conditions, e.g. bad weather or weak lighting, while LiDAR scanners are too expensive to get widely deployed in commercial applications. Radar has been drawing more and more attention due to its robustness and low cost. In this paper, we propose a scene-aware radar learning framework for accurate and robust object detection. First, the learning framework contains branches conditioning on the scene category of the radar sequence; with each branch optimized for a specific type of scene. Second, three different 3D autoencoder-based architectures are proposed for radar object detection and ensemble learning is performed over the different architectures to further boost the final performance. Third, we propose novel scene-aware sequence mix augmentation (SceneMix) and scene-specific post-processing to generate more robust detection results. In the ROD2021 Challenge, we achieved a final result of average precision of 75.0% and an average recall of 81.0%. Moreover, in the parking lot scene, our framework ranks first with an average precision of 97.8% and an average recall of 98.6%, which demonstrates the effectiveness of our framework.

Corvin Deboeser, Jordan Ivanchev, Thomas Braud et al.

This paper introduces the SEAD framework that simplifies the process of designing and describing autonomous vehicle platooning manoeuvres. Although a large body of research has been formulating platooning manoeuvres, it is still challenging to design, describe, read, and understand them. This difficulty largely arises from missing formalisation. To fill this gap, we analysed existing ways of describing manoeuvres, derived the causes of difficulty, and designed a framework that simplifies the manoeuvre design process. Alongside, a Manoeuvre Design Language was developed to structurally describe manoeuvres in a machine-readable format. Unlike state-of-the-art manoeuvre descriptions that require one state machine for every participating vehicle, the SEAD framework allows describing any manoeuvre from the single perspective of the platoon leader. %As a proof of concept, the proposed framework was implemented in the mixed traffic simulation environment BEHAVE for an autonomous highway scenario. Using this framework, we implemented several manoeuvres as they were described in literature. To demonstrate the applicability of the framework, an experiment was performed to evaluate the execution time performance of multiple alternatives of the Join-Middle manoeuvre. This proof-of-concept experiment revealed that the manoeuvre execution time can be reduced by 28 \% through parallelising various steps without considerable secondary effects. We hope that the SEAD framework will pave the way for further research in the area of new manoeuvre design and optimisation by largely simplifying and unifying platooning manoeuvre representation.

Xiangyu Yue, Zangwei Zheng, Shanghang Zhang et al.

Unsupervised Domain Adaptation (UDA) transfers predictive models from a fully-labeled source domain to an unlabeled target domain. In some applications, however, it is expensive even to collect labels in the source domain, making most previous works impractical. To cope with this problem, recent work performed instance-wise cross-domain self-supervised learning, followed by an additional fine-tuning stage. However, the instance-wise self-supervised learning only learns and aligns low-level discriminative features. In this paper, we propose an end-to-end Prototypical Cross-domain Self-Supervised Learning (PCS) framework for Few-shot Unsupervised Domain Adaptation (FUDA). PCS not only performs cross-domain low-level feature alignment, but it also encodes and aligns semantic structures in the shared embedding space across domains. Our framework captures category-wise semantic structures of the data by in-domain prototypical contrastive learning; and performs feature alignment through cross-domain prototypical self-supervision. Compared with state-of-the-art methods, PCS improves the mean classification accuracy over different domain pairs on FUDA by 10.5%, 3.5%, 9.0%, and 13.2% on Office, Office-Home, VisDA-2017, and DomainNet, respectively. Our project page is at http://xyue.io/pcs-fuda/index.html

Jay Shenoy, Edward Kim, Xiangyu Yue et al.

Safely interacting with humans is a significant challenge for autonomous driving. The performance of this interaction depends on machine learning-based modules of an autopilot, such as perception, behavior prediction, and planning. These modules require training datasets with high-quality labels and a diverse range of realistic dynamic behaviors. Consequently, training such modules to handle rare scenarios is difficult because they are, by definition, rarely represented in real-world datasets. Hence, there is a practical need to augment datasets with synthetic data covering these rare scenarios. In this paper, we present a platform to model dynamic and interactive scenarios, generate the scenarios in simulation with different modalities of labeled sensor data, and collect this information for data augmentation. To our knowledge, this is the first integrated platform for these tasks specialized to the autonomous driving domain.

Sicheng Zhao, Xuanbai Chen, Xiangyu Yue et al.

Thanks to large-scale labeled training data, deep neural networks (DNNs) have obtained remarkable success in many vision and multimedia tasks. However, because of the presence of domain shift, the learned knowledge of the well-trained DNNs cannot be well generalized to new domains or datasets that have few labels. Unsupervised domain adaptation (UDA) studies the problem of transferring models trained on one labeled source domain to another unlabeled target domain. In this paper, we focus on UDA in visual emotion analysis for both emotion distribution learning and dominant emotion classification. Specifically, we design a novel end-to-end cycle-consistent adversarial model, termed CycleEmotionGAN++. First, we generate an adapted domain to align the source and target domains on the pixel-level by improving CycleGAN with a multi-scale structured cycle-consistency loss. During the image translation, we propose a dynamic emotional semantic consistency loss to preserve the emotion labels of the source images. Second, we train a transferable task classifier on the adapted domain with feature-level alignment between the adapted and target domains. We conduct extensive UDA experiments on the Flickr-LDL & Twitter-LDL datasets for distribution learning and ArtPhoto & FI datasets for emotion classification. The results demonstrate the significant improvements yielded by the proposed CycleEmotionGAN++ as compared to state-of-the-art UDA approaches.

Daniel J. Fremont, Edward Kim, Tommaso Dreossi et al.

We propose a new probabilistic programming language for the design and analysis of cyber-physical systems, especially those based on machine learning. Specifically, we consider the problems of training a system to be robust to rare events, testing its performance under different conditions, and debugging failures. We show how a probabilistic programming language can help address these problems by specifying distributions encoding interesting types of inputs, then sampling these to generate specialized training and test data. More generally, such languages can be used to write environment models, an essential prerequisite to any formal analysis. In this paper, we focus on systems like autonomous cars and robots, whose environment at any point in time is a 'scene', a configuration of physical objects and agents. We design a domain-specific language, Scenic, for describing scenarios that are distributions over scenes and the behaviors of their agents over time. As a probabilistic programming language, Scenic allows assigning distributions to features of the scene, as well as declaratively imposing hard and soft constraints over the scene. We develop specialized techniques for sampling from the resulting distribution, taking advantage of the structure provided by Scenic's domain-specific syntax. Finally, we apply Scenic in a case study on a convolutional neural network designed to detect cars in road images, improving its performance beyond that achieved by state-of-the-art synthetic data generation methods.

Sicheng Zhao, Xiangyu Yue, Shanghang Zhang et al.

Large-scale labeled training datasets have enabled deep neural networks to excel across a wide range of benchmark vision tasks. However, in many applications, it is prohibitively expensive and time-consuming to obtain large quantities of labeled data. To cope with limited labeled training data, many have attempted to directly apply models trained on a large-scale labeled source domain to another sparsely labeled or unlabeled target domain. Unfortunately, direct transfer across domains often performs poorly due to the presence of domain shift or dataset bias. Domain adaptation is a machine learning paradigm that aims to learn a model from a source domain that can perform well on a different (but related) target domain. In this paper, we review the latest single-source deep unsupervised domain adaptation methods focused on visual tasks and discuss new perspectives for future research. We begin with the definitions of different domain adaptation strategies and the descriptions of existing benchmark datasets. We then summarize and compare different categories of single-source unsupervised domain adaptation methods, including discrepancy-based methods, adversarial discriminative methods, adversarial generative methods, and self-supervision-based methods. Finally, we discuss future research directions with challenges and possible solutions.

Yingshui Tan, Baihong Jin, Qiushi Cui et al.

Incipient anomalies present milder symptoms compared to severe ones, and are more difficult to detect and diagnose due to their close resemblance to normal operating conditions. The lack of incipient anomaly examples in the training data can pose severe risks to anomaly detection methods that are built upon Machine Learning (ML) techniques, because these anomalies can be easily mistaken as normal operating conditions. To address this challenge, we propose to utilize the uncertainty information available from ensemble learning to identify potential misclassified incipient anomalies. We show in this paper that ensemble learning methods can give improved performance on incipient anomalies and identify common pitfalls in these models through extensive experiments on two real-world datasets. Then, we discuss how to design more effective ensemble models for detecting incipient anomalies.

Baihong Jin, Yingshui Tan, Albert Liu et al.

Incipient anomalies present milder symptoms compared to severe ones, and are more difficult to detect and diagnose due to their close resemblance to normal operating conditions. The lack of incipient anomaly examples in the training data can pose severe risks to anomaly detection methods that are built upon Machine Learning (ML) techniques, because these anomalies can be easily mistaken as normal operating conditions. To address this challenge, we propose to utilize the uncertainty information available from ensemble learning to identify potential misclassified incipient anomalies. We show in this paper that ensemble learning methods can give improved performance on incipient anomalies and identify common pitfalls in these models through extensive experiments on two real-world datasets. Then, we discuss how to design more effective ensemble models for detecting incipient anomalies.

Yingshui Tan, Baihong Jin, Xiangyu Yue et al.

Ensemble learning is widely applied in Machine Learning (ML) to improve model performance and to mitigate decision risks. In this approach, predictions from a diverse set of learners are combined to obtain a joint decision. Recently, various methods have been explored in literature for estimating decision uncertainties using ensemble learning; however, determining which metrics are a better fit for certain decision-making applications remains a challenging task. In this paper, we study the following key research question in the selection of uncertainty metrics: when does an uncertainty metric outperforms another? We answer this question via a rigorous analysis of two commonly used uncertainty metrics in ensemble learning, namely ensemble mean and ensemble variance. We show that, under mild assumptions on the ensemble learners, ensemble mean is preferable with respect to ensemble variance as an uncertainty metric for decision making. We empirically validate our assumptions and theoretical results via an extensive case study: the diagnosis of referable diabetic retinopathy.

Baihong Jin, Yingshui Tan, Yuxin Chen et al.

Intermediate-Severity (IS) faults present milder symptoms compared to severe faults, and are more difficult to detect and diagnose due to their close resemblance to normal operating conditions. The lack of IS fault examples in the training data can pose severe risks to Fault Detection and Diagnosis (FDD) methods that are built upon Machine Learning (ML) techniques, because these faults can be easily mistaken as normal operating conditions. Ensemble models are widely applied in ML and are considered promising methods for detecting out-of-distribution (OOD) data. We identify common pitfalls in these models through extensive experiments with several popular ensemble models on two real-world datasets. Then, we discuss how to design more effective ensemble models for detecting and diagnosing IS faults.

Baihong Jin, Yingshui Tan, Yuxin Chen et al.

The Monte Carlo dropout method has proved to be a scalable and easy-to-use approach for estimating the uncertainty of deep neural network predictions. This approach was recently applied to Fault Detection and Di-agnosis (FDD) applications to improve the classification performance on incipient faults. In this paper, we propose a novel approach of augmenting the classification model with an additional unsupervised learning task. We justify our choice of algorithm design via an information-theoretical analysis. Our experimental results on three datasets from diverse application domains show that the proposed method leads to improved fault detection and diagnosis performance, especially on out-of-distribution examples including both incipient and unknown faults.

Xiangyu Yue, Yang Zhang, Sicheng Zhao et al.

We propose to harness the potential of simulation for the semantic segmentation of real-world self-driving scenes in a domain generalization fashion. The segmentation network is trained without any data of target domains and tested on the unseen target domains. To this end, we propose a new approach of domain randomization and pyramid consistency to learn a model with high generalizability. First, we propose to randomize the synthetic images with the styles of real images in terms of visual appearances using auxiliary datasets, in order to effectively learn domain-invariant representations. Second, we further enforce pyramid consistency across different "stylized" images and within an image, in order to learn domain-invariant and scale-invariant features, respectively. Extensive experiments are conducted on the generalization from GTA and SYNTHIA to Cityscapes, BDDS and Mapillary; and our method achieves superior results over the state-of-the-art techniques. Remarkably, our generalization results are on par with or even better than those obtained by state-of-the-art simulation-to-real domain adaptation methods, which access the target domain data at training time.

Baihong Jin, Yingshui Tan, Alexander Nettekoven et al.

We present a novel unsupervised deep learning approach that utilizes the encoder-decoder architecture for detecting anomalies in sequential sensor data collected during industrial manufacturing. Our approach is designed not only to detect whether there exists an anomaly at a given time step, but also to predict what will happen next in the (sequential) process. We demonstrate our approach on a dataset collected from a real-world testbed. The dataset contains images collected under both normal conditions and synthetic anomalies. We show that the encoder-decoder model is able to identify the injected anomalies in a modern manufacturing process in an unsupervised fashion. In addition, it also gives hints about the temperature non-uniformity of the testbed during manufacturing, which is what we are not aware of before doing the experiment.

Tommaso Dreossi, Shromona Ghosh, Alberto Sangiovanni-Vincentelli et al.

Deep neural networks have been shown to lack robustness to small input perturbations. The process of generating the perturbations that expose the lack of robustness of neural networks is known as adversarial input generation. This process depends on the goals and capabilities of the adversary, In this paper, we propose a unifying formalization of the adversarial input generation process from a formal methods perspective. We provide a definition of robustness that is general enough to capture different formulations. The expressiveness of our formalization is shown by modeling and comparing a variety of adversarial attack techniques.

Pan Li, Baihong Jin, Ruoxuan Xiong et al.

We present a machine learning approach to the solution of chance constrained optimizations in the context of voltage regulation problems in power system operation. The novelty of our approach resides in approximating the feasible region of uncertainty with an ellipsoid. We formulate this problem using a learning model similar to Support Vector Machines (SVM) and propose a sampling algorithm that efficiently trains the model. We demonstrate our approach on a voltage regulation problem using standard IEEE distribution test feeders.

Baihong Jin, Yuxin Chen, Dan Li et al.

It is important to identify the change point of a system's health status, which usually signifies an incipient fault under development. The One-Class Support Vector Machine (OC-SVM) is a popular machine learning model for anomaly detection and hence could be used for identifying change points; however, it is sometimes difficult to obtain a good OC-SVM model that can be used on sensor measurement time series to identify the change points in system health status. In this paper, we propose a novel approach for calibrating OC-SVM models. The approach uses a heuristic search method to find a good set of input data and hyperparameters that yield a well-performing model. Our results on the C-MAPSS dataset demonstrate that OC-SVM can also achieve satisfactory accuracy in detecting change point in time series with fewer training data, compared to state-of-the-art deep learning approaches. In our case study, the OC-SVM calibrated by the proposed model is shown to be useful especially in scenarios with limited amount of training data.

Tommaso Dreossi, Shromona Ghosh, Xiangyu Yue et al.

We present a novel framework for augmenting data sets for machine learning based on counterexamples. Counterexamples are misclassified examples that have important properties for retraining and improving the model. Key components of our framework include a counterexample generator, which produces data items that are misclassified by the model and error tables, a novel data structure that stores information pertaining to misclassifications. Error tables can be used to explain the model's vulnerabilities and are used to efficiently generate counterexamples for augmentation. We show the efficacy of the proposed framework by comparing it to classical augmentation techniques on a case study of object detection in autonomous driving based on deep neural networks.

Xiangyu Yue, Bichen Wu, Sanjit A. Seshia et al.

3D LiDAR scanners are playing an increasingly important role in autonomous driving as they can generate depth information of the environment. However, creating large 3D LiDAR point cloud datasets with point-level labels requires a significant amount of manual annotation. This jeopardizes the efficient development of supervised deep learning algorithms which are often data-hungry. We present a framework to rapidly create point clouds with accurate point-level labels from a computer game. The framework supports data collection from both auto-driving scenes and user-configured scenes. Point clouds from auto-driving scenes can be used as training data for deep learning algorithms, while point clouds from user-configured scenes can be used to systematically test the vulnerability of a neural network, and use the falsifying examples to make the neural network more robust through retraining. In addition, the scene images can be captured simultaneously in order for sensor fusion tasks, with a method proposed to do automatic calibration between the point clouds and captured scene images. We show a significant improvement in accuracy (+9%) in point cloud segmentation by augmenting the training dataset with the generated synthesized data. Our experiments also show by testing and retraining the network using point clouds from user-configured scenes, the weakness/blind spots of the neural network can be fixed.

Marcell Vazquez-Chanlatte, Shromona Ghosh, Jyotirmoy V. Deshmukh et al.

Cyber-physical systems of today are generating large volumes of time-series data. As manual inspection of such data is not tractable, the need for learning methods to help discover logical structure in the data has increased. We propose a logic-based framework that allows domain-specific knowledge to be embedded into formulas in a parametric logical specification over time-series data. The key idea is to then map a time series to a surface in the parameter space of the formula. Given this mapping, we identify the Hausdorff distance between boundaries as a natural distance metric between two time-series data under the lens of the parametric specification. This enables embedding non-trivial domain-specific knowledge into the distance metric and then using off-the-shelf machine learning tools to label the data. After labeling the data, we demonstrate how to extract a logical specification for each label. Finally, we showcase our technique on real world traffic data to learn classifiers/monitors for slow-downs and traffic jams.

Somil Bansal, Shromona Ghosh, Alberto Sangiovanni-Vincentelli et al.

With an increasing use of data-driven models to control robotic systems, it has become important to develop a methodology for validating such models before they can be deployed to design a controller for the actual system. Specifically, it must be ensured that the controller designed for a learned model would perform as expected on the actual physical system. We propose a context-specific validation framework to quantify the quality of a learned model based on a distance measure between the closed-loop actual system and the learned model. We then propose an active sampling scheme to compute a probabilistic upper bound on this distance in a sample-efficient manner. The proposed framework validates the learned model against only those behaviors of the system that are relevant for the purpose for which we intend to use this model, and does not require any a priori knowledge of the system dynamics. Several simulations illustrate the practicality of the proposed framework for validating the models of real-world systems, and consequently, for controller synthesis.

Tommaso Dreossi, Shromona Ghosh, Alberto Sangiovanni-Vincentelli et al.

We present a framework to systematically analyze convolutional neural networks (CNNs) used in classification of cars in autonomous vehicles. Our analysis procedure comprises an image generator that produces synthetic pictures by sampling in a lower dimension image modification subspace and a suite of visualization tools. The image generator produces images which can be used to test the CNN and hence expose its vulnerabilities. The presented framework can be used to extract insights of the CNN classifier, compare across classification models, or generate training and validation datasets.

Marcell Vazquez-Chanlatte, Shromona Ghosh, Vasumathi Raman et al.

We address the problem of synthesizing reactive controllers for cyber-physical systems subject to Signal Temporal Logic (STL) specifications in the presence of adversarial inputs. Given a finite horizon, we define a reactive hierarchy of control problems that differ in the degree of information available to the system about the adversary's actions over the horizon. We show how to construct reactive controllers at various levels of the hierarchy, leveraging the existence of Lipschitz bounds on system dynamics and the quantitative semantics of STL. Our approach, a counterexample-guided inductive synthesis (CEGIS) scheme based on optimization and satisfiability modulo theories (SMT) solving, builds a strategy tree representing the interaction between the system and its environment. In every iteration of the CEGIS loop, we use a mix of optimization and SMT to maximally discard controllers falsified by a given counterexample. Our approach can be applied to any system with local Lipschitz-bounded dynamics, including linear, piecewise-linear and differentially-flat systems. Finally we show an application in the autonomous car domain.

Jiwei Li, Pierluigi Nuzzo, Alberto Sangiovanni-Vincentelli et al.

We develop an assume-guarantee contract framework for the design of cyber-physical systems, modeled as closed-loop control systems, under probabilistic requirements. We use a variant of signal temporal logic, namely, Stochastic Signal Temporal Logic (StSTL) to specify system behaviors as well as contract assumptions and guarantees, thus enabling automatic reasoning about requirements of stochastic systems. Given a stochastic linear system representation and a set of requirements captured by bounded StSTL contracts, we propose algorithms that can check contract compatibility, consistency, and refinement, and generate a controller to guarantee that a contract is satisfied, following a stochastic model predictive control approach. Our algorithms leverage encodings of the verification and control synthesis tasks into mixed integer optimization problems, and conservative approximations of probabilistic constraints that produce both sound and tractable problem formulations. We illustrate the effectiveness of our approach on a few examples, including the design of embedded controllers for aircraft power distribution networks.

Yasser Shoukry, Pierluigi Nuzzo, Nicola Bezzo et al.

We address the problem of estimating the state of a differentially flat system from measurements that may be corrupted by an adversarial attack. In cyber-physical systems, malicious attacks can directly compromise the system's sensors or manipulate the communication between sensors and controllers. We consider attacks that only corrupt a subset of sensor measurements. We show that the possibility of reconstructing the state under such attacks is characterized by a suitable generalization of the notion of s-sparse observability, previously introduced by some of the authors in the linear case. We also extend our previous work on the use of Satisfiability Modulo Theory solvers to estimate the state under sensor attacks to the context of differentially flat systems. The effectiveness of our approach is illustrated on the problem of controlling a quadrotor under sensor attacks.

Yasser Shoukry, Pierluigi Nuzzo, Alberto Puggelli et al.

We address the problem of detecting and mitigating the effect of malicious attacks to the sensors of a linear dynamical system. We develop a novel, efficient algorithm that uses a Satisfiability-Modulo-Theory approach to isolate the compromised sensors and estimate the system state despite the presence of the attack, thus harnessing the intrinsic combinatorial complexity of the problem. By leveraging results from formal methods over real numbers, we provide guarantees on the soundness and completeness of our algorithm. We then report simulation results to compare its runtime performance with alternative techniques. Finally, we demonstrate its application to the problem of controlling an unmanned ground vehicle.

Pierluigi Nuzzo, John Finn, Mohammad Mozumdar et al.

In an aircraft electric power system (EPS), a supervisory control unit must actuate a set of switches to distribute power from generators to loads, while satisfying safety, reliability and real-time performance requirements. To reduce expensive re-design steps in current design methodologies, such a control problem is generally addressed based on minor incremental changes on top of consolidated solutions, since it is difficult to estimate the impact of earlier design decisions on the final implementation. In this paper, we introduce a methodology for the design space exploration and virtual prototyping of EPS supervisory control protocols, following the platform-based design (PBD) paradigm. Moreover, we describe the modeling infrastructure that supports the methodology. In PBD, design space exploration is carried out as a sequence of refinement steps from the initial specification towards a final implementation, by mapping higher-level behavioral models into a set of library components at a lower level of abstraction. In our flow, the system specification is captured using SysML requirement and structure diagrams. State-machine diagrams enable verification of the control protocol at a high level of abstraction, while lowerlevel hybrid models, implemented in Simulink, are used to verify properties related to physical quantities, such as time, voltage and current values. The effectiveness of our approach is illustrated on a prototype EPS control protocol design.