Model-based deep reinforcement learning has achieved success in various domains that require high sample efficiencies, such as Go and robotics. However, there are some remaining issues, such as planning efficient explorations to learn more accurate dynamic models, evaluating the uncertainty of the learned models, and more rational utilization of models. To mitigate these issues, we present MEEE, a model-ensemble method that consists of optimistic exploration and weighted exploitation. During exploration, unlike prior methods directly selecting the optimal action that maximizes the expected accumulative return, our agent first generates a set of action candidates and then seeks out the optimal action that takes both expected return and future observation novelty into account. During exploitation, different discounted weights are assigned to imagined transition tuples according to their model uncertainty respectively, which will prevent model predictive error propagation in agent training. Experiments on several challenging continuous control benchmark tasks demonstrated that our approach outperforms other model-free and model-based state-of-the-art methods, especially in sample complexity.

We introduce the problem of sleeping dueling bandits with stochastic preferences and adversarial availabilities (DB-SPAA). In almost all dueling bandit applications, the decision space often changes over time; eg, retail store management, online shopping, restaurant recommendation, search engine optimization, etc. Surprisingly, this `sleeping aspect' of dueling bandits has never been studied in the literature. Like dueling bandits, the goal is to compete with the best arm by sequentially querying the preference feedback of item pairs. The non-triviality however results due to the non-stationary item spaces that allow any arbitrary subsets items to go unavailable every round. The goal is to find an optimal `no-regret' policy that can identify the best available item at each round, as opposed to the standard `fixed best-arm regret objective' of dueling bandits. We first derive an instance-specific lower bound for DB-SPAA $\Omega( \sum_{i =1}^{K-1}\sum_{j=i+1}^K \frac{\log T}{\Delta(i,j)})$, where $K$ is the number of items and $\Delta(i,j)$ is the gap between items $i$ and $j$. This indicates that the sleeping problem with preference feedback is inherently more difficult than that for classical multi-armed bandits (MAB). We then propose two algorithms, with near optimal regret guarantees. Our results are corroborated empirically.

Bayesian probability theory is the fundamental framework for dealing with uncertain data, and is at the core of practical systems in machine learning and robotics [23, 11]. A probabilistic model relates unknown variables of interest to observable, known or assumed quantities and most generally takes the form of a graph whose connections encode those relationships. Inference is the process of forming the posterior distribution to determine properties of the unknown variables, given the observations, such as their most probable values or their full marginal distributions. There are various possible algorithms for probabilistic inference, many of which take advantage of specific problem structure for fast performance. Efficient inference on models represented by large, dynamic and highly inter-connected graphs however remains computationally challenging and is already a limiting factor in real embodied systems.

A prominent technique for reducing the memory footprint of Spiking Neural Networks (SNNs) without decreasing the accuracy significantly is quantization. However, the state-of-the-art only focus on employing the weight quantization directly from a specific quantization scheme, i.e., either the post-training quantization (PTQ) or the in-training quantization (ITQ), and do not consider (1) quantizing other SNN parameters (e.g., neuron membrane potential), (2) exploring different combinations of quantization approaches (i.e., quantization schemes, precision levels, and rounding schemes), and (3) selecting the SNN model with a good memory-accuracy trade-off at the end. Therefore, the memory saving offered by these state-of-the-art to meet the targeted accuracy is limited, thereby hindering processing SNNs on the resource-constrained systems (e.g., the IoT-Edge devices). Towards this, we propose Q-SpiNN, a novel quantization framework for memory-efficient SNNs. The key mechanisms of the Q-SpiNN are: (1) employing quantization for different SNN parameters based on their significance to the accuracy, (2) exploring different combinations of quantization schemes, precision levels, and rounding schemes to find efficient SNN model candidates, and (3) developing an algorithm that quantifies the benefit of the memory-accuracy trade-off obtained by the candidates, and selects the Pareto-optimal one. The experimental results show that, for the unsupervised network, the Q-SpiNN reduces the memory footprint by ca. 4x, while maintaining the accuracy within 1% from the baseline on the MNIST dataset. For the supervised network, the Q-SpiNN reduces the memory by ca. 2x, while keeping the accuracy within 2% from the baseline on the DVS-Gesture dataset.

Deployed machine learning models are confronted with the problem of changing data over time, a phenomenon also called concept drift. While existing approaches of concept drift detection already show convincing results, they require true labels as a prerequisite for successful drift detection. Especially in many real-world application scenarios--like the ones covered in this work--true labels are scarce, and their acquisition is expensive. Therefore, we introduce a new algorithm for drift detection, Uncertainty Drift Detection (UDD), which is able to detect drifts without access to true labels. Our approach is based on the uncertainty estimates provided by a deep neural network in combination with Monte Carlo Dropout. Structural changes over time are detected by applying the ADWIN technique on the uncertainty estimates, and detected drifts trigger a retraining of the prediction model. In contrast to input data-based drift detection, our approach considers the effects of the current input data on the properties of the prediction model rather than detecting change on the input data only (which can lead to unnecessary retrainings). We show that UDD outperforms other state-of-the-art strategies on two synthetic as well as ten real-world data sets for both regression and classification tasks.

ABSTRACT We explore the potential to control terrain vehicles using deep reinforcement in scenarios where human operators and traditional control methods are inadequate. This letter presents a controller that perceives, plans, and successfully controls a 16-tonne forestry vehicle with two frame articulation joints, six wheels, and their actively articulated suspensions to traverse rough terrain. The carefully shaped reward signal promotes safe, environmental, and efficient driving, which leads to the emergence of unprecedented driving skills. We test learned skills in a virtual environment, including terrains reconstructed from high-density laser scans of forest sites. The results confirm that deep reinforcement learning has the potential to enhance control of vehicles with complex dynamics and high-dimensional observation data compared to human operators or traditional control methods, especially in rough terrain. 1 INTRODUCTION Deep reinforcement learning has recently shown promise for locomotion tasks, but its usefulness to learn control of heavy vehicles in rough terrain is widely unknown. Conventionally, the design of rough terrain vehicles strives to promote high traversability and be easily operated by humans. The drivelines involve differentials and bogie suspension that provide ground compliance and reduces the many degrees of freedom, leaving only speed and heading for the operator to control. An attractive alternative is to use actively articulated suspensions and individual wheel control. These have the potential to reduce the energy consumption and ground damage, yet increase traversability and tip over stability [11, 6, 21, 10, 9].

There is an old saying, "to err is human". Is humanity imposing double standards on artificial intelligence? COVID-19 has wreaked havoc of magnitude proportion across the globe, not even machines had been spared of its ramifications. This is exemplified by an incident back in April 2020 when traders attempted to sell off May's oil future contracts which resulted in a futures oil prices nosedive into negative digits (Note: these are crude oil we are speaking of, so normal consumers will have little use for such oil – you cannot store it in your backyard either). On 20 April 2020, the price of West Texas Intermediate futures contracts traded to as low as negative $40.32 per barrel.

The energy sector faces pressing challenges and needs to act with urgency. Policy commitments to a net-zero future, such as the Paris Agreement, mean the transformation to a low-carbon economy must come at pace. Major disruption to the electricity sector is on the cards as governments ramp up renewables and transition away from fossil fuels. While renewable energy looks set to flourish amid this backdrop, its intermittent nature means solutions will need to be found to keep grids stable. Additionally, the industry is changing from a market based on commodity pricing to a market based on technology solutions in order to integrate renewable energy.

Fractures are widely developed in hydrocarbon reservoirs and constitute the accumulation spaces and transport channels of oil and gas. Fracture detection is a fundamental task for reservoir characterization. From prestack seismic gathers, anisotropic analysis and inversion were commonly applied to characterize the dominant orientations and relative intensities of fractures. However, the existing methods were mostly based on the vertical aligned facture hypothesis, it is impossible for them to recognize fracture dip. Furthermore, it is difficult or impractical for existing methods to attain the real fracture densities. Based on data-driven deep learning, this paper designed a convolutional neural network to perform prestack fracture detection. Capitalizing on the connections between seismic responses and fracture parameters, a suitable azimuth dataset was firstly generated through fracture effective medium modeling and anisotropic plane wave analyzing. Then a multi-input and multi-output convolutional neural network was constructed to simultaneously detect fracture density, dip and strike azimuth. The application on a practical survey validated the effectiveness of the proposed CNN model.

Smart grids are part of a growing "smart" phenomenon involving distributed devices that are wirelessly connected and intelligently controlled to automate decisions normally left to people. The Internet of Things (IoT) is the most popular example of this trend, with smart phones, thermostats, fridges, and even cars working in concert to share real-time data and make decisions autonomously. Smart grid technology does the same thing – but for energy. This comprehensive guide explains how smart electrical grids work, why they are important, and how they are helping to revolutionize the electricity landscape – especially as distributed energy sources (DERs) like solar, wind, and battery storage continue to place stress on America's aging power infrastructure. You may also enjoy this brief 30-minute podcast that introduces the challenges of smart grids and highlights some of the benefits of AI to improve energy and utility operations.