This article develops a deep reinforcement learning (Deep-RL) framework for dynamic pricing on managed lanes with multiple access locations and heterogeneity in travelers' value of time, origin, and destination. This framework relaxes assumptions in the literature by considering multiple origins and destinations, multiple access locations to the managed lane, en route diversion of travelers, partial observability of the sensor readings, and stochastic demand and observations. The problem is formulated as a partially observable Markov decision process (POMDP) and policy gradient methods are used to determine tolls as a function of real-time observations. Tolls are modeled as continuous and stochastic variables, and are determined using a feedforward neural network. The method is compared against a feedback control method used for dynamic pricing. We show that Deep-RL is effective in learning toll policies for maximizing revenue, minimizing total system travel time, and other joint weighted objectives, when tested on real-world transportation networks. The Deep-RL toll policies outperform the feedback control heuristic for the revenue maximization objective by generating revenues up to 9.5% higher than the heuristic and for the objective minimizing total system travel time (TSTT) by generating TSTT up to 10.4% lower than the heuristic. We also propose reward shaping methods for the POMDP to overcome the undesired behavior of toll policies, like the jam-and-harvest behavior of revenue-maximizing policies. Additionally, we test transferability of the algorithm trained on one set of inputs for new input distributions and offer recommendations on real-time implementations of Deep-RL algorithms. The source code for our experiments is available online at https://github.com/venktesh22/ExpressLanes_Deep-RL

Finite-horizon sequential decision problems arise naturally in many machine learning contexts; examples include Bayesian optimization and Bayesian quadrature. Computing the optimal policy for such problems requires solving Bellman equations, which are generally intractable. Most existing work resorts to myopic approximations by limiting the horizon to only a single time-step, which can perform poorly in balancing exploration and exploitation. We propose a general framework for efficient, nonmyopic approximation of the optimal policy by drawing a connection between the optimal adaptive policy and its non-adaptive counterpart. Our proposal is to compute an optimal batch of points, then select a single point from within this batch to evaluate. We realize this idea for both Bayesian optimization and Bayesian quadrature and demonstrate that our proposed method significantly outperforms common myopic alternatives on a variety of tasks.

The characterization of amorphous materials is especially challenging because their lack of long-range order makes it difficult to define structural metrics. In this work, we apply deep learning algorithms to accurately classify amorphous materials and characterize their structural features. Specifically, we show that convolutional neural networks and message passing neural networks can classify two-dimensional liquids and liquid-cooled glasses from molecular dynamics simulations with greater than 0.98 AUC, with no a priori assumptions about local particle relationships, even when the liquids and glasses are prepared at the same inherent structure energy. Furthermore, we demonstrate that message passing neural networks surpass convolutional neural networks in this context in both accuracy and interpretability. We extract a clear interpretation of how message passing neural networks evaluate liquid and glass structures by using a self-attention mechanism. Using this interpretation, we derive three novel structural metrics that accurately characterize glass formation. The methods presented here provide us with a procedure to identify important structural features in materials that could be missed by standard techniques and give us a unique insight into how these neural networks process data. I. INTRODUCTION Classifying material structures and predicting their properties are important tasks in materials science. The behavior of materials often depends strongly on their underlying structure, and understanding these structure-property relationships relies on accurately describing the structural features of a material. However, quantifying structure-property relationships and identifying structural features in complex materials are difficult tasks. A variety of standard techniques have been developed to analyze material structures. Some of the most common techniques include the Steinhardt bond order parameters, 1 Bond Angle Analysis (BAA), 2 and Common Neighbor Analysis (CNA), 3 which are useful for detecting order-disorder transitions and differentiating between crystal structures in ordered samples. As discussed in Reinhardt et al., 4 the Steinhardt bond order parameters can be stymied by thermal fluctuations or am-a) Electronic mail: swansonk1@uchicago.edu BAA relies on a small set of crystalline reference structures that may not be present in amorphous samples. CNA is more flexible than BAA, but it cannot provide accurate information about particles that do not exhibit known symmetries, making analysis of irregular structures challenging.

Generating forecasts for time series with multiple seasonal cycles is an important use-case for many industries nowadays. Accounting for the multi-seasonal patterns becomes necessary to generate more accurate and meaningful forecasts in these contexts. In this paper, we propose Long Short-Term Memory Multi-Seasonal Net (LSTM-MSNet), a decompositionbased, unified prediction framework to forecast time series with multiple seasonal patterns. The current state of the art in this space are typically univariate methods, in which the model parameters of each time series are estimated independently. Consequently, these models are unable to include key patterns and structures that may be shared by a collection of time series. In contrast, LSTM-MSNet is a globally trained Long Short-Term Memory network (LSTM), where a single prediction model is built across all the available time series to exploit the crossseries knowledge in a group of related time series. Furthermore, our methodology combines a series of state-of-the-art multiseasonal decomposition techniques to supplement the LSTM learning procedure. In our experiments, we are able to show that on datasets from disparate data sources, like e.g. the popular M4 forecasting competition, a decomposition step is beneficial, whereas in the common real-world situation of homogeneous series from a single application, exogenous seasonal variables or no seasonal preprocessing at all are better choices. All options are readily included in the framework and allow us to achieve competitive results for both cases, outperforming many state-ofthe-art multi-seasonal forecasting methods

Vision-based grasping systems typically adopt an open-loop execution of a planned grasp. This policy can fail due to many reasons, including ubiquitous calibration error. Recovery from a failed grasp is further complicated by visual occlusion, as the hand is usually occluding the vision sensor as it attempts another open-loop regrasp. This work presents MAT, a tactile closed-loop method capable of realizing grasps provided by a coarse initial positioning of the hand above an object. Our algorithm is a deep reinforcement learning (RL) policy optimized through the clipped surrogate objective within a maximum entropy RL framework to balance exploitation and exploration. The method utilizes tactile and proprioceptive information to act through both fine finger motions and larger regrasp movements to execute stable grasps. A novel curriculum of action motion magnitude makes learning more tractable and helps turn common failure cases into successes. Careful selection of features that exhibit small sim-to-real gaps enables this tactile grasping policy, trained purely in simulation, to transfer well to real world environments without the need for additional learning. Experimentally, this methodology improves over a vision-only grasp success rate substantially on a multi-fingered robot hand. When this methodology is used to realize grasps from coarse initial positions provided by a vision-only planner, the system is made dramatically more robust to calibration errors in the camera-robot transform.

SunPower has launched a web application that can design rooftop solar projects in seconds. SunPower's Design Studio combines SunPowers's Instant Design technology, Google Cloud and Google Sunroof to deliver customizable residential solar designs that take into account roof size, shading and energy potential. Homeowners can modify their design based on their energy needs, resulting in more accurate predictions of electricity bill and energy savings. "With SunPower Design Studio, we've created a new solar buying experience," said Jake Wachman, digital VP for SunPower. "We're making solar accessible by enabling homeowners nationwide to envision solar on their home and understand savings at lightning-fast speeds. With SunPower Design Studio, we're changing how homeowners go solar."

Founded in London in 2010 and acquired by Google in 2014, AI company DeepMind Technologies Ltd. reportedly reduced the amount of energy required to cool Google's data centers by 40 percent. DeepMind reported these results in July 2016, however, the company claims that it first began applying machine learning two years prior to improve energy usage. Specifically, a set of data center operating scenarios and parameters were used to train a system of neural networks. The neural network "learned" how the data center functioned and began identifying opportunities for optimization. Google claims that data was pulled from thousands of sensors located in the data centers.

As they steer through the fast-evolving environment, oil and gas companies are continuing their digital transformation journeys with a view to drive growth, productivity, efficiency, and safety across their operations. As the industry's adoption of technologies continues to advanced, their efforts to explore and forge new business models has also developed accordingly. Oil and gas companies worldwide are going through a widespread change. On the one hand, the considerable price volatility as shifting geopolitical dynamics sees the supply-and-demand equations diverge in various geographies, and on the other, it is necessary to comply with increasing environmental regulations designed to support de-carbonization. Meanwhile, an escalating inclination for oil and gas companies is to create partnerships and collaborative arrangements in areas like supply chain integration, logistics, trading, and payments.

The future looks bright for data-driven building operations. These solutions can save time when operating a building and deliver better outcomes for occupants. One difficulty with understanding AI in buildings is the fact that it can be applied in a variety of ways. There is a range of realistic use cases. The breadth of the technology occasionally creates confusion about what AI actually can do.

How likely is it that a few initial node activations are amplified to produce large response cascades that span a considerable part of an entire network? Our answer to this question relies on the Independent Cascade Model for weighted directed networks. In using this model, most of our insights have been derived from the study of average effects. Here, we shift the focus on the full probability distribution of the final cascade size. This shift allows us to explore both typical cascade outcomes and improbable but relevant extreme events. We present an efficient message passing algorithm to compute the final cascade size distribution and activation probabilities of nodes conditional on the final cascade size. Our approach is exact on trees but can be applied to any network topology. It approximates locally treelike networks well and can lead to surprisingly good performance on more dense networks, as we show using real world data, including a miRNAmiRNA probabilistic interaction network for gastrointestinal cancer. We demonstrate the utility of our algorithms for clustering of nodes according to their functionality and influence maximization. Introduction The Independent Cascade Model (ICM) is a cornerstone in the study of spreading processes on networks. Many related optimization algorithms require sampling from the model.