Showing papers by "Alberto Sangiovanni-Vincentelli published in 2020"

A Review of Single-Source Deep Unsupervised Visual Domain Adaptation.

Abstract: 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 (DA) 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 article, we review the latest single-source deep unsupervised DA methods focused on visual tasks and discuss new perspectives for future research. We begin with the definitions of different DA strategies and the descriptions of existing benchmark datasets. We then summarize and compare different categories of single-source unsupervised DA 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.

A Review of Single-Source Deep Unsupervised Visual Domain Adaptation

Abstract: 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.

Optimized Selection of Reliable and Cost-Effective Safety-Critical System Architectures

Abstract: We address the problem of synthesizing safety-critical embedded and cyber-physical system architectures to minimize a cost function while guaranteeing the desired reliability. We represent a system architecture as a configurable graph in which both the nodes (components) and edges (interconnections) may fail. We then propose a compact analytical formalism to efficiently reason about the reliability of the overall system based on the failure probabilities of the components, and provide expressions of the design constraints that avoid exhaustive enumeration of failure cases on all possible graph configurations. Based on these constraints, we cast the synthesis problem as an optimization problem and propose monolithic and iterative optimization schemes to decrease the problem complexity. We implement the proposed algorithms in the ArchEx framework, leveraging a pattern-based specification language to facilitate problem formulation. Design problems from aircraft electric power distribution networks and reconfigurable industrial manufacturing systems illustrate the effectiveness of our approach.

Gordian: Formal Reasoning-based Outlier Detection for Secure Localization

Abstract: Accurate localization from Cyber-Physical Systems (CPS) is a critical enabling technology for context-aware applications and control As localization plays an increasingly safety-critical role, location systems must be able to identify and eliminate faulty measurements to prevent dangerously inaccurate localization In this article, we consider the range-based localization problem and propose a method to detect coordinated adversarial corruption on anchor positions and distance measurements Our algorithm, Gordian, rapidly finds attacks by identifying geometric inconsistencies at the graph level without requiring assumptions about hardware, ranging mechanisms, or cryptographic protocols We give necessary conditions for which attack detection is guaranteed to be successful in the noiseless case, and we use that intuition to extend Gordian to the noisy case where fewer guarantees are possible In simulations generated from real-world sensor noise, we empirically show that Gordian’s trilateration counterexample generation procedure enables rapid attack detection even for combinatorially difficult problems

Emotional Semantics-Preserved and Feature-Aligned CycleGAN for Visual Emotion Adaptation

Abstract: 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.

Industrial Control via Application Containers:Maintaining determinism in IAAS

Abstract: Industry 4.0 is changing fundamentally data collection, its storage and analysis in industrial processes, enabling novel application such as flexible manufacturing of highly customized products. Real-time control of these processes, however, has not yet realized its full potential in using the collected data to drive further development. Indeed, typical industrial control systems are tailored to the plant they need to control, making reuse and adaptation a challenge. In the past, the need to solve plant specific problems overshadowed the benefits of physically isolating a control system from its plant. We believe that modern virtualization techniques, specifically application containers, present a unique opportunity to decouple control from plants. This separation permits us to fully realize the potential for highly distributed, and transferable industrial processes even with real-time constraints arising from time-critical sub-processes. In this paper, we explore the challenges and opportunities of shifting industrial control software from dedicated hardware to bare-metal servers or (edge) cloud computing platforms using off-the-shelf technology. We present a migration architecture and show, using a specifically developed orchestration tool, that containerized applications can run on shared resources without compromising scheduled execution within given time constraints. Through latency and computational performance experiments we explore limits of three system setups and summarize lessons learned.

Exploiting Uncertainties from Ensemble Learners to Improve Decision-Making in Healthcare AI.

Abstract: 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.

Using Ensemble Classifiers to Detect Incipient Anomalies.

Abstract: 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

Are Ensemble Classifiers Powerful Enough for the Detection and Diagnosis of Intermediate-Severity Faults?

Abstract: 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.

ODRE Workshop: Probabilistic Dynamic Hard Real-Time Scheduling in HPC

Abstract: Industry 4.0 is changing fundamentally the way data is collected, stored and analyzed in industrial processes. While this change enables novel application such as flexible manufacturing of highly customized products, the real-time control of these processes, however, has not yet realized its full potential. We believe that modern virtualization techniques, specifically application containers, present a unique opportunity to decouple control functionality from associated hardware. Through it, we can fully realize the potential for highly distributed and transferable industrial processes even with real-time constraints arising from time-critical sub-processes. In this paper, we present a specifically developed orchestration tool to manage the challenges and opportunities of shifting industrial control software from dedicated hardware to bare-metal servers or (edge) cloud computing platforms. Using off-the-shelf technology, the proposed tool can manage the execution of containerized applications on shared resources without compromising hard real-time execution determinism. Through first experimental results, we confirm the viability and analyzed the behavior of resource shared systems with strict real-time requirements. We then describe experiments set out to deliver expected results and gather performance, application scope and limits of the presented approach.

The Quotient in Preorder Theories

Abstract: Seeking the largest solution to an expression of the form A x <= B is a common task in several domains of engineering and computer science. This largest solution is commonly called quotient. Across domains, the meanings of the binary operation and the preorder are quite different, yet the syntax for computing the largest solution is remarkably similar. This paper is about finding a common framework to reason about quotients. We only assume we operate on a preorder endowed with an abstract monotonic multiplication and an involution. We provide a condition, called admissibility, which guarantees the existence of the quotient, and which yields its closed form. We call preordered heaps those structures satisfying the admissibility condition. We show that many existing theories in computer science are preordered heaps, and we are thus able to derive a quotient for them, subsuming existing solutions when available in the literature. We introduce the concept of sieved heaps to deal with structures which are given over multiple domains of definition. We show that sieved heaps also have well-defined quotients.

The Quotient in Preorder Theories

Abstract: Seeking the largest solution to an expression of the form A x <= B is a common task in several domains of engineering and computer science. This largest solution is commonly called quotient. Across domains, the meanings of the binary operation and the preorder are quite different, yet the syntax for computing the largest solution is remarkably similar. This paper is about finding a common framework to reason about quotients. We only assume we operate on a preorder endowed with an abstract monotonic multiplication and an involution. We provide a condition, called admissibility, which guarantees the existence of the quotient, and which yields its closed form. We call preordered heaps those structures satisfying the admissibility condition. We show that many existing theories in computer science are preordered heaps, and we are thus able to derive a quotient for them, subsuming existing solutions when available in the literature. We introduce the concept of sieved heaps to deal with structures which are given over multiple domains of definition. We show that sieved heaps also have well-defined quotients.

A Customizable Dynamic Scenario Modeling and Data Generation Platform for Autonomous Driving.

Abstract: 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.

Generalizing Fault Detection Against Domain Shifts Using Stratification-Aware Cross-Validation.

Abstract: 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