Many industrial sectors have been collecting big sensor data. With recent technologies for processing big data, companies can exploit this for automatic failure detection and prevention. We propose the first completely automated method for failure analysis, machine-learning fault trees from raw observational data with continuous variables. Our method scales well and is tested on a real-world, five-year dataset of domestic heater operations in The Netherlands, with 31 million unique heater-day readings, each containing 27 sensor and 11 failure variables. Our method builds on two previous procedures: the C4.5 decision-tree learning algorithm, and the LIFT fault tree learning algorithm from Boolean data. C4.5 pre-processes each continuous variable: it learns an optimal numerical threshold which distinguishes between faulty and normal operation of the top-level system. These thresholds discretise the variables, thus allowing LIFT to learn fault trees which model the root failure mechanisms of the system and are explainable. We obtain fault trees for the 11 failure variables, and evaluate them in two ways: quantitatively, with a significance score, and qualitatively, with domain specialists. Some of the fault trees learnt have almost maximum significance (above 0.95), while others have medium-to-low significance (around 0.30), reflecting the difficulty of learning from big, noisy, real-world sensor data. The domain specialists confirm that the fault trees model meaningful relationships among the variables.

In recent years, considerable advancements have been made in the area of Generative Adversarial Networks (GANs), particularly with the advent of style-based architectures that address many key shortcomings - both in terms of modeling capabilities and network interpretability. Despite these improvements, the adoption of such approaches in the domain of satellite imagery is not straightforward. Typical vision datasets used in generative tasks are well-aligned and annotated, and exhibit limited variability. In contrast, satellite imagery exhibits great spatial and spectral variability, wide presence of fine, high-frequency details, while the tedious nature of annotating satellite imagery leads to annotation scarcity - further motivating developments in unsupervised learning. In this light, we present the first pre-trained style- and wavelet-based GAN model that can readily synthesize a wide gamut of realistic satellite images in a variety of settings and conditions - while also preserving high-frequency information. Furthermore, we show that by analyzing the intermediate activations of our network, one can discover a multitude of interpretable semantic directions that facilitate the guided synthesis of satellite images in terms of high-level concepts (e.g., urbanization) without using any form of supervision. Via a set of qualitative and quantitative experiments we demonstrate the efficacy of our framework, in terms of suitability for downstream tasks (e.g., data augmentation), quality of synthetic imagery, as well as generalization capabilities to unseen datasets.

Aarti Singh, an associate professor in the Machine Learning Department, will use her research on decision-making algorithms to study the evolution of the universe as part of the Simons Collaboration on Learning the Universe. This international collaboration includes researchers from CMU, Columbia University, Harvard University, Princeton University, Lawrence Berkeley National Labs, the Flatiron Institute and international partners from Canada, France, Germany and Sweden. For scientists to understand how the universe evolved, they must know its initial conditions and the physical laws governing those conditions. Since these aren't knowable, they can only be inferred through observation. The collaboration -- directed by Greg Bryan, a professor of astronomy at Columbia University, and made possible by the Simons Foundation -- will repeatedly select sets of initial conditions, predict how they would be observed now, compare that to real observations of galaxies and gas, and then compute the likelihood of those initial conditions.

Ottonomy robots help navigate businesses with staffing shortages for retail and restaurant industries. Our fully autonomous robots can deliver food & beverages, groceries, and packages to curbside, last mile, and even indoor environments. Ottonomy robots are available on a "RaaS" (Robotics as a Service) model. Our business customers get access to a quicker, safer, and more economical delivery option as compared to traditional 3rd party delivery services. Above all these robots are set to reduce carbon emissions and improve quality of life.

Cyber-physical systems (CPSs) are smart systems that depend on the synergy of cyber and physical components. They link the physical world (e.g. through sensors, actuators, robotics, and embedded systems) with the virtual world of information processing. Applications of CPS have the tremendous potential of improving convenience, comfort, and safety in our daily life. This paper provides a brief introduction to CPSs and their applications. The term "cyber-physical system" (CPS) was coined in 2006 by Helen Gill of the US National Science Foundation (Henshaw, 2016).

SPAN develops products that accelerate the rapid adoption of renewable energy in the home. The flagship SPAN Smart Panel is the first true evolution for the traditional home electric panel, harnessing enhanced technology for metering, monitoring, and control. An expanded product suite of intelligent, integrated solutions radically lowers the cost and complexity of energy upgrades–including solar, batteries and EVs–empowering homeowners to be active, resilient and informed players in the energy market. We are seeking a senior data engineer to join our team building the cloud based glue that gives our users access to the rich information and controls provided by the SPAN Panel. Our system collects a large volume of energy monitoring data that needs to be stored, processed, and exposed in different ways for different end users.

Ever since the world has stepped forward towards the age of digitalization, things have never been the same. From the introduction of the internet to the expansion of the mobile-first concept and innovations like artificial intelligence and machine learning, people have experienced the highest exposure to technology ever. Amidst all this development and expansion, one thing that has scaled dynamically is Artificial Intelligence. From the expansion of neural networks to energy use, data sets, and the prevalence of society, the growth of AI has made way for significant ethical concerns. Before we jump on unraveling the most common ethical issues surrounding artificial intelligence, let us begin with developing an understanding of what ethical AI is.

We propose an algorithm to (i) learn online a deep signed distance function (SDF) with a LiDAR-equipped robot to represent the 3D environment geometry, and (ii) plan collision-free trajectories given this deep learned map. Our algorithm takes a stream of incoming LiDAR scans and continually optimizes a neural network to represent the SDF of the environment around its current vicinity. When the SDF network quality saturates, we cache a copy of the network, along with a learned confidence metric, and initialize a new SDF network to continue mapping new regions of the environment. We then concatenate all the cached local SDFs through a confidence-weighted scheme to give a global SDF for planning. For planning, we make use of a sequential convex model predictive control (MPC) algorithm. The MPC planner optimizes a dynamically feasible trajectory for the robot while enforcing no collisions with obstacles mapped in the global SDF. We show that our online mapping algorithm produces higher-quality maps than existing methods for online SDF training. In the WeBots simulator, we further showcase the combined mapper and planner running online -- navigating autonomously and without collisions in an unknown environment.

Mapping near-field pollutant concentration is essential to track accidental toxic plume dispersion in urban areas. By solving a large part of the turbulence spectrum, large-eddy simulations (LES) have the potential to accurately represent pollutant concentration spatial variability. Finding a way to synthesize this large amount of information to improve the accuracy of lower-fidelity operational models (e.g. providing better turbulence closure terms) is particularly appealing. This is a challenge in multi-query contexts, where LES become prohibitively costly to deploy to understand how plume flow and tracer dispersion change with various atmospheric and source parameters. To overcome this issue, we propose a non-intrusive reduced-order model combining proper orthogonal decomposition (POD) and Gaussian process regression (GPR) to predict LES field statistics of interest associated with tracer concentrations. GPR hyperpararameters are optimized component-by-component through a maximum a posteriori (MAP) procedure informed by POD. We provide a detailed analysis of the reducedorder model performance on a two-dimensional case study corresponding to a turbulent atmospheric boundary-layer flow over a surface-mounted obstacle. We show that near-source concentration heterogeneities upstream of the obstacle require a large number of POD modes to be well captured. We also show that the component-by-component optimization allows to capture the range of spatial scales in the POD modes, especially the shorter concentration patterns in the high-order modes. The reduced-order model predictions remain acceptable if the learning database is made of at least fifty to hundred LES snapshot providing a first estimation of the required budget to move towards more realistic atmospheric dispersion applications.

Entailment trees have been proposed to simulate the human reasoning process of explanation generation in the context of open--domain textual question answering. However, in practice, manually constructing these explanation trees proves a laborious process that requires active human involvement. Given the complexity of capturing the line of reasoning from question to the answer or from claim to premises, the issue arises of how to assist the user in efficiently constructing multi--level entailment trees given a large set of available facts. In this paper, we frame the construction of entailment trees as a sequence of active premise selection steps, i.e., for each intermediate node in an explanation tree, the expert needs to annotate positive and negative examples of premise facts from a large candidate list. We then iteratively fine--tune pre--trained Transformer models with the resulting positive and tightly controlled negative samples and aim to balance the encoding of semantic relationships and explanatory entailment relationships. Experimental evaluation confirms the measurable efficiency gains of the proposed active fine--tuning method in facilitating entailment trees construction: up to 20\% improvement in explanatory premise selection when compared against several alternatives.