Meta-gradients provide a general approach for optimizing the meta-parameters of reinforcement learning (RL) algorithms. Estimation of meta-gradients is central to the performance of these meta-algorithms, and has been studied in the setting of MAML-style short-horizon meta-RL problems. In this context, prior work has investigated the estimation of the Hessian of the RL objective, as well as tackling the problem of credit assignment to pre-adaptation behavior by making a sampling correction. However, we show that Hessian estimation, implemented for example by DiCE and its variants, always adds bias and can also add variance to meta-gradient estimation. Meanwhile, meta-gradient estimation has been studied less in the important long-horizon setting, where backpropagation through the full inner optimization trajectories is not feasible. We study the bias and variance tradeoff arising from truncated backpropagation and sampling correction, and additionally compare to evolution strategies, which is a recently popular alternative strategy to long-horizon meta-learning. While prior work implicitly chooses points in this bias-variance space, we disentangle the sources of bias and variance and present an empirical study that relates existing estimators to each other.

The field of electricity price forecasting has seen significant advances in the last years, including the development of new, more accurate forecast models. These models leverage statistical relationships in previously observed data to predict the future; however, there is a lack of analysis explaining these models, which limits their real world applicability in critical infrastructure. In this paper, using data from the Belgian electricity markets, we explore a state-of-the-art forecasting model to understand if its predictions can be trusted in more general settings than the limited context it is trained in. If the model produces poor predictions in extreme conditions or if its predictions are inconsistent with reality, it cannot be relied upon in real-world where these forecasts are used in downstream decision-making activities. Our results show that, despite being largely accurate enough in general, even state of the art forecasts struggle with remaining consistent with reality.

Many real-life optimization problems frequently contain one or more constraints or objectives for which there are no explicit formulas. If data is however available, these data can be used to learn the constraints. The benefits of this approach are clearly seen, however there is a need for this process to be carried out in a structured manner. This paper therefore provides a framework for Optimization with Constraint Learning (OCL) which we believe will help to formalize and direct the process of learning constraints from data. This framework includes the following steps: (i) setup of the conceptual optimization model, (ii) data gathering and preprocessing, (iii) selection and training of predictive models, (iv) resolution of the optimization model, and (v) verification and improvement of the optimization model. We then review the recent OCL literature in light of this framework, and highlight current trends, as well as areas for future research.

Neural style transfer is a powerful computer vision technique that can incorporate the artistic "style" of one image to the "content" of another. The underlying theory behind the approach relies on the assumption that the style of an image is represented by the Gram matrix of its features, which is typically extracted from pre-trained convolutional neural networks (e.g., VGG-19). This idea does not straightforwardly extend to time series stylization since notions of style for two-dimensional images are not analogous to notions of style for one-dimensional time series. In this work, a novel formulation of time series style transfer is proposed for the purpose of synthetic data generation and enhancement. We introduce the concept of stylized features for time series, which is directly related to the time series realism properties, and propose a novel stylization algorithm, called StyleTime, that uses explicit feature extraction techniques to combine the underlying content (trend) of one time series with the style (distributional properties) of another. Further, we discuss evaluation metrics, and compare our work to existing state-of-the-art time series generation and augmentation schemes. To validate the effectiveness of our methods, we use stylized synthetic data as a means for data augmentation to improve the performance of recurrent neural network models on several forecasting tasks.

Taking advantage of the many recent advances in deep learning, text-to-image generative models currently have the merit of attracting the general public attention. Two of these models, DALL-E 2 and Imagen, have demonstrated that highly photorealistic images could be generated from a simple textual description of an image. Based on a novel approach for image generation called diffusion models, text-to-image models enable the production of many different types of high resolution images, where human imagination is the only limit. However, these models require exceptionally large amounts of computational resources to train, as well as handling huge datasets collected from the internet. In addition, neither the codebase nor the models have been released. It consequently prevents the AI community from experimenting with these cutting-edge models, making the reproduction of their results complicated, if not impossible. In this thesis, we aim to contribute by firstly reviewing the different approaches and techniques used by these models, and then by proposing our own implementation of a text-to-image model. Highly based on DALL-E 2, we introduce several slight modifications to tackle the high computational cost induced. We thus have the opportunity to experiment in order to understand what these models are capable of, especially in a low resource regime. In particular, we provide additional and analyses deeper than the ones performed by the authors of DALL-E 2, including ablation studies. Besides, diffusion models use so-called guidance methods to help the generating process. We introduce a new guidance method which can be used in conjunction with other guidance methods to improve image quality. Finally, the images generated by our model are of reasonably good quality, without having to sustain the significant training costs of state-of-the-art text-to-image models.

This paper presents a compact system, OpenPneu, to support the pneumatic actuation for multi-chambers on soft robots. Micro-pumps are employed in the system to generate airflow and therefore no extra input as compressed air is needed. Our system conducts modular design to provide good scalability, which has been demonstrated on a prototype with ten air channels. Each air channel of OpenPneu is equipped with both the inflation and the deflation functions to provide a full range pressure supply from positive to negative with a maximal flow rate at 1.7 L/min. High precision closed-loop control of pressures has been built into our system to achieve stable and efficient dynamic performance in actuation. An open-source control interface and API in Python are provided. We also demonstrate the functionality of OpenPneu on three soft robotic systems with up to 10 chambers.

The effective representation, precessing, analysis, and visualization of large-scale structured data over graphs are gaining a lot of attention. So far most of the literature has focused on real-valued signals. However, signals are often sparse in the Fourier domain, and more informative and compact representations for them can be obtained using the complex envelope of their spectral components, as opposed to the original real-valued signals. Motivated by this fact, in this work we generalize graph convolutional neural networks (GCN) to the complex domain, deriving the theory that allows to incorporate a complex-valued graph shift operators (GSO) in the definition of graph filters (GF) and process complex-valued graph signals (GS). The theory developed can handle spatio-temporal complex network processes. We prove that complex-valued GCNs are stable with respect to perturbations of the underlying graph support, the bound of the transfer error and the bound of error propagation through multiply layers. Then we apply complex GCN to power grid state forecasting, power grid cyber-attack detection and localization.

Deep Reinforcement Learning (DRL) is employed to develop autonomously optimized and custom-designed heat-treatment processes that are both, microstructure-sensitive and energy efficient. Different from conventional supervised machine learning, DRL does not rely on static neural network training from data alone, but a learning agent autonomously develops optimal solutions, based on reward and penalty elements, with reduced or no supervision. In our approach, a temperature-dependent Allen-Cahn model for phase transformation is used as the environment for the DRL agent, serving as the model world in which it gains experience and takes autonomous decisions. The agent of the DRL algorithm is controlling the temperature of the system, as a model furnace for heat-treatment of alloys. Microstructure goals are defined for the agent based on the desired microstructure of the phases. After training, the agent can generate temperature-time profiles for a variety of initial microstructure states to reach the final desired microstructure state. The agent's performance and the physical meaning of the heat-treatment profiles generated are investigated in detail. In particular, the agent is capable of controlling the temperature to reach the desired microstructure starting from a variety of initial conditions. This capability of the agent in handling a variety of conditions paves the way for using such an approach also for recycling-oriented heat treatment process design where the initial composition can vary from batch to batch, due to impurity intrusion, and also for the design of energy-efficient heat treatments. For testing this hypothesis, an agent without penalty on the total consumed energy is compared with one that considers energy costs. The energy cost penalty is imposed as an additional criterion on the agent for finding the optimal temperature-time profile.

There's a new use for 5G networks: spotting wildfires before they get out of control, an increasing worry as climate change makes fires spread faster and burn longer than before. The startup Pano AI uses a series of cameras that survey the wilderness and AI algorithms that watch for telltale smoke -- an indicator of small blazes that could grow into raging wildfires. That footage is sent to the startup's headquarters for human confirmation, and if a fire is burning, evidence is sent to clients who could be affected. While Pano AI had been sending evidence photos over 4G LTE networks at slow rates of around 20 to 30 6-megapixel images per minute, its new partnership with T-Mobile has it using the carrier's 5G network to send video at 30 frames per second, which is around 90 times more data. Ultimately, getting evidence to Pano AI's clients, which include utility companies, much quicker on 5G means a faster response from firefighters and potentially squashing big fires before they get dangerous.

The lesser spotted eagle is endangered in Germany.Hinze, K/DPA via ZUMA Press This story was originally published by the Guardian and is reproduced here as part of the Climate Desk collaboration. Small in size, sensitive of constitution and with only 130 breeding pairs surviving locally in the wild, the lesser spotted eagle of the Oder delta lives up to its name. In Germany, key questions over the country's energy future hang on the question of whether artificial intelligence systems can do a better job of spotting the reclusive animal than birdwatchers do. Lesser spotted eagles (named after the drop-shaped spots on their feathers) are fond of riding thermals over many of the flatlands earmarked for a mass expansion of onshore windfarms by a German government under pressure to compensate for a pending loss of nuclear power, coal plants and Russian gas. Because lesser spotted eagles in mid-flight are unused to vertical obstacles, and keep their eyes focused on mice, lizard or frog-shaped prey below, conservationists say, they are known to occasionally collide with the rotor blades of wind turbines.