With the widespread adoption of machine learning (ML), series of models (i.e., where the outputs of predictive models are used as inputs to separate predictive models) are increasingly common. Examples include: (1) stacked generalization, a widely used technique [1-5] to improve generalization performance by ensembling the predictions of many models (called base-learners) using another model (called a meta-learner) [6], (2) neural network feature extraction, where models are trained on features extracted using neural networks [7, 8], typically for structured data [9-11], and (3) consumer scores, where predictive models that describe a specific behavior (e.g., credit scores [12]) are used as inputs to downstream predictive models. For example, a bank may use a model to predict customers' loan eligibility on the basis of their bank statements and their credit score, which itself is often a predictive model [13]. Explaining a series of models is crucial for debugging and building trust, even more so because a series of models is inherently harder to explain compared to a single model. One popular paradigm for explaining models are local feature attributions, which explain why a model makes a prediction for a single sample (known as the "explicand" [14]). Existing model-agnostic local feature attribution methods (e.g., IME [15], LIME [16], KernelSHAP [17]) work regardless of the specific model being explained. They can explain a series of models, but suffer from two distinct shortcomings: (1) their sampling-based estimates of feature importance are inherently variable, and (2) they have high computational cost which may not be tractable for large pipelines.

3D object detection often involves complicated training and testing pipelines, which require substantial domain knowledge about individual datasets. Inspired by recent non-maximum suppression-free 2D object detection models, we propose a 3D object detection architecture on point clouds. Our method models 3D object detection as message passing on a dynamic graph, generalizing the DGCNN framework to predict a set of objects. In our construction, we remove the necessity of post-processing via object confidence aggregation or non-maximum suppression. To facilitate object detection from sparse point clouds, we also propose a set-to-set distillation approach customized to 3D detection. This approach aligns the outputs of the teacher model and the student model in a permutation-invariant fashion, significantly simplifying knowledge distillation for the 3D detection task. Our method achieves state-of-the-art performance on autonomous driving benchmarks. We also provide abundant analysis of the detection model and distillation framework.

Within the field of process mining, several different trace clustering approaches exist for partitioning traces or process instances into similar groups. Typically, this partitioning is based on certain patterns or similarity between the traces, or driven by the discovery of a process model for each cluster. The main drawback of these techniques, however, is that their solutions are usually hard to evaluate or justify by domain experts. In this paper, we present two constrained trace clustering techniques that are capable to leverage expert knowledge in the form of instance-level constraints. In an extensive experimental evaluation using two real-life datasets, we show that our novel techniques are indeed capable of producing clustering solutions that are more justifiable without a substantial negative impact on their quality.

Autonomous cyber-physical systems (CPS) can improve safety and efficiency for safety-critical applications, but require rigorous testing before deployment. The complexity of these systems often precludes the use of formal verification and real-world testing can be too dangerous during development. Therefore, simulation-based techniques have been developed that treat the system under test as a black box operating in a simulated environment. Safety validation tasks include finding disturbances in the environment that cause the system to fail (falsification), finding the most-likely failure, and estimating the probability that the system fails. Motivated by the prevalence of safety-critical artificial intelligence, this work provides a survey of state-of-the-art safety validation techniques for CPS with a focus on applied algorithms and their modifications for the safety validation problem. We present and discuss algorithms in the domains of optimization, path planning, reinforcement learning, and importance sampling. Problem decomposition techniques are presented to help scale algorithms to large state spaces, which are common for CPS. A brief overview of safety-critical applications is given, including autonomous vehicles and aircraft collision avoidance systems. Finally, we present a survey of existing academic and commercially available safety validation tools.

E-Commerce marketplaces support millions of daily transactions, and some disagreements between buyers and sellers are unavoidable. Resolving disputes in an accurate, fast, and fair manner is of great importance for maintaining a trustworthy platform. Simple cases can be automated, but intricate cases are not sufficiently addressed by hard-coded rules, and therefore most disputes are currently resolved by people. In this work we take a first step towards automatically assisting human agents in dispute resolution at scale. We construct a large dataset of disputes from the eBay online marketplace, and identify several interesting behavioral and linguistic patterns. We then train classifiers to predict dispute outcomes with high accuracy. We explore the model and the dataset, reporting interesting correlations, important features, and insights.

Given an unlabeled dataset, wherein we have access only to pairwise similarities (or distances), how can we effectively (1) detect outliers, and (2) annotate/tag the outliers by type? Outlier detection has a large literature, yet we find a key gap in the field: to our knowledge, no existing work addresses the outlier annotation problem. Outliers are broadly classified into 3 types, representing distinct patterns that could be valuable to analysts: (a) global outliers are severe yet isolate cases that do not repeat, e.g., a data collection error; (b) local outliers diverge from their peers within a context, e.g., a particularly short basketball player; and (c) collective outliers are isolated micro-clusters that may indicate coalition or repetitions, e.g., frauds that exploit the same loophole. This paper presents C-AllOut: a novel and effective outlier detector that annotates outliers by type. It is parameter-free and scalable, besides working only with pairwise similarities (or distances) when it is needed. We show that C-AllOut achieves on par or significantly better performance than state-of-the-art detectors when spotting outliers regardless of their type. It is also highly effective in annotating outliers of particular types, a task that none of the baselines can perform.

We introduce Ego4D, a massive-scale egocentric video dataset and benchmark suite. It offers 3,025 hours of daily-life activity video spanning hundreds of scenarios (household, outdoor, workplace, leisure, etc.) captured by 855 unique camera wearers from 74 worldwide locations and 9 different countries. The approach to collection is designed to uphold rigorous privacy and ethics standards with consenting participants and robust de-identification procedures where relevant. Ego4D dramatically expands the volume of diverse egocentric video footage publicly available to the research community. Portions of the video are accompanied by audio, 3D meshes of the environment, eye gaze, stereo, and/or synchronized videos from multiple egocentric cameras at the same event. Furthermore, we present a host of new benchmark challenges centered around understanding the first-person visual experience in the past (querying an episodic memory), present (analyzing hand-object manipulation, audio-visual conversation, and social interactions), and future (forecasting activities). By publicly sharing this massive annotated dataset and benchmark suite, we aim to push the frontier of first-person perception. Project page: https://ego4d-data.org/

Continuous blood pressure (BP) measurements can reflect a body's response to diseases and serve as a predictor of cardiovascular and other health conditions. While current cuff-based BP measurement methods are incapable of providing continuous BP readings, invasive BP monitoring methods also tend to cause patient dissatisfaction and can potentially cause infection. In this research, we developed a method for estimating blood pressure based on the features extracted from Electrocardiogram (ECG) and Photoplethysmogram (PPG) signals and the Arterial Blood Pressure (ABP) data. The vector of features extracted from the preprocessed ECG and PPG signals is used in this approach, which include Pulse Transit Time (PTT), PPG Intensity Ratio (PIR), and Heart Rate (HR), as the input of a clustering algorithm and then developing separate regression models like Random Forest Regression, Gradient Boosting Regression, and Multilayer Perceptron Regression algorithms for each resulting cluster. We evaluated and compared the findings to create the model with the highest accuracy by applying the clustering approach and identifying the optimal number of clusters, and eventually the acceptable prediction model. The paper compares the results obtained with and without this clustering. The results show that the proposed clustering approach helps obtain more accurate estimates of Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP). Given the inconsistency, high dispersion, and multitude of trends in the datasets for different features, using the clustering approach improved the estimation accuracy by 50-60%.

Many-to-one maps are ubiquitous in machine learning, from the image recognition model that assigns a multitude of distinct images to the concept of "cat" to the time series forecasting model which assigns a range of distinct time-series to a single scalar regression value. While the primary use of such models is naturally to associate correct output to each input, in many problems it is also useful to be able to explore, understand, and sample from a model's fibers, which are the set of input values $x$ such that $f(x) = y$, for fixed $y$ in the output space. In this paper we show that popular generative architectures are ill-suited to such tasks. Motivated by this we introduce a novel generative architecture, a Bundle Network, based on the concept of a fiber bundle from (differential) topology. BundleNets exploit the idea of a local trivialization wherein a space can be locally decomposed into a product space that cleanly encodes the many-to-one nature of the map. By enforcing this decomposition in BundleNets and by utilizing state-of-the-art invertible components, investigating a network's fibers becomes natural.

As the primary cause of software defects, human error is the key to understanding, and perhaps to predicting and avoiding them. Little research has been done to predict defects on the basis of the cognitive errors that cause them. This paper proposes an approach to predicting software defects through knowledge about the cognitive mechanisms of human errors. Our theory is that the main process behind a software defect is that an error-prone scenario triggers human error modes, which psychologists have observed to recur across diverse activities. Software defects can then be predicted by identifying such scenarios, guided by this knowledge of typical error modes. The proposed idea emphasizes predicting the exact location and form of a possible defect. We conducted two case studies to demonstrate and validate this approach, with 55 programmers in a programming competition and 5 analysts serving as the users of the approach. We found it impressive that the approach was able to predict, at the requirement phase, the exact locations and forms of 7 out of the 22 (31.8%) specific types of defects that were found in the code. The defects predicted tended to be common defects: their occurrences constituted 75.7% of the total number of defects in the 55 developed programs; each of them was introduced by at least two persons. The fraction of the defects introduced by a programmer that were predicted was on average (over all programmers) 75%. Furthermore, these predicted defects were highly persistent through the debugging process. If the prediction had been used to successfully prevent these defects, this could have saved 46.2% of the debugging iterations. This excellent capability of forecasting the exact locations and forms of possible defects at the early phases of software development recommends the approach for substantial benefits to defect prevention and early detection.