Specifying tasks with videos is a powerful technique towards acquiring novel and general robot skills. However, reasoning over mechanics and dexterous interactions can make it challenging to scale learning contact-rich manipulation. In this work, we focus on the problem of visual non-prehensile planar manipulation: given a video of an object in planar motion, find contact-aware robot actions that reproduce the same object motion. We propose a novel architecture, Differentiable Learning for Manipulation (\ours), that combines video decoding neural models with priors from contact mechanics by leveraging differentiable optimization and finite difference based simulation. Through extensive simulated experiments, we investigate the interplay between traditional model-based techniques and modern deep learning approaches. We find that our modular and fully differentiable architecture performs better than learning-only methods on unseen objects and motions. \url{https://github.com/baceituno/dlm}.

The Internet of Things (IoT) is on the verge of a major paradigm shift. In the IoT system of the future, IoFT, the cloud will be substituted by the crowd where model training is brought to the edge, allowing IoT devices to collaboratively extract knowledge and build smart analytics/models while keeping their personal data stored locally. This paradigm shift was set into motion by the tremendous increase in computational power on IoT devices and the recent advances in decentralized and privacy-preserving model training, coined as federated learning (FL). This article provides a vision for IoFT and a systematic overview of current efforts towards realizing this vision. Specifically, we first introduce the defining characteristics of IoFT and discuss FL data-driven approaches, opportunities, and challenges that allow decentralized inference within three dimensions: (i) a global model that maximizes utility across all IoT devices, (ii) a personalized model that borrows strengths across all devices yet retains its own model, (iii) a meta-learning model that quickly adapts to new devices or learning tasks. We end by describing the vision and challenges of IoFT in reshaping different industries through the lens of domain experts. Those industries include manufacturing, transportation, energy, healthcare, quality & reliability, business, and computing.

Keyword spotting (KWS) has become an indispensable part of many intelligent devices surrounding us, as audio is one of the most efficient ways of interacting with these devices. The accuracy and performance of KWS solutions have been the main focus of the researchers, and thanks to deep learning, substantial progress has been made in this domain. However, as the use of KWS spreads into IoT devices, energy efficiency becomes a very critical requirement besides the performance. We believe KWS solutions that would seek power optimization both in the hardware and the neural network (NN) model architecture are advantageous over many solutions in the literature where mostly the architecture side of the problem is considered. In this work, we designed an optimized KWS CNN model by considering end-to-end energy efficiency for the deployment at MAX78000, an ultra-low-power CNN accelerator. With the combined hardware and model optimization approach, we achieve 96.3\% accuracy for 12 classes while only consuming 251 uJ per inference. We compare our results with other small-footprint neural network-based KWS solutions in the literature. Additionally, we share the energy consumption of our model in power-optimized ARM Cortex-M4F to depict the effectiveness of the chosen hardware for the sake of clarity.

A digital replica of Earth could help scientists better model the future of our planet and find solutions to problems wrought by climate change. The advanced model, dubbed Digital Twin Earth, is being developed by the European Space Agency (ESA) and its partners based on data and images from Earth-observation satellites and sensors on the ground. To run reliably, the project will require new advanced artificial intelligence algorithms and powerful supercomputers, which are currently being developed. ESA and its partners discussed their progress in the runup to the UN Climate Change Conference COP26, a two-week event that's currently taking place in Glasgow, Scotland. ESA launched the Digital Twin Earth project in 2020 and invited researchers and tech companies from across Europe to present their progress during an event called PhiWeek, which took place Oct. 11 to Oct. 15.

We present a careful comparison of two model-free control algorithms, Evolution Strategies (ES) and Proximal Policy Optimization (PPO), with receding horizon model predictive control (MPC) for operating simulated, price responsive water heaters. Four MPC variants are considered: a one-shot controller with perfect forecasting yielding optimal control; a limited-horizon controller with perfect forecasting; a mean forecasting-based controller; and a two-stage stochastic programming controller using historical scenarios. In all cases, the MPC model for water temperature and electricity price are exact; only water demand is uncertain. For comparison, both ES and PPO learn neural network-based policies by directly interacting with the simulated environment under the same scenarios used by MPC. All methods are then evaluated on a separate one-week continuation of the demand time series. We demonstrate that optimal control for this problem is challenging, requiring more than 8-hour lookahead for MPC with perfect forecasting to attain the minimum cost. Despite this challenge, both ES and PPO learn good general purpose policies that outperform mean forecast and two-stage stochastic MPC controllers in terms of average cost and are more than two orders of magnitude faster at computing actions. We show that ES in particular can leverage parallelism to learn a policy in under 90 seconds using 1150 CPU cores.

Many engineering problems require the prediction of realization-to-realization variability or a refined description of modeled quantities. In that case, it is necessary to sample elements from unknown high-dimensional spaces with possibly millions of degrees of freedom. While there exist methods able to sample elements from probability density functions (PDF) with known shapes, several approximations need to be made when the distribution is unknown. In this paper the sampling method, as well as the inference of the underlying distribution, are both handled with a data-driven method known as generative adversarial networks (GAN), which trains two competing neural networks to produce a network that can effectively generate samples from the training set distribution. In practice, it is often necessary to draw samples from conditional distributions. When the conditional variables are continuous, only one (if any) data point corresponding to a particular value of a conditioning variable may be available, which is not sufficient to estimate the conditional distribution. This work handles this problem using an a priori estimation of the conditional moments of a PDF. Two approaches, stochastic estimation, and an external neural network are compared here for computing these moments; however, any preferred method can be used. The algorithm is demonstrated in the case of the deconvolution of a filtered turbulent flow field. It is shown that all the versions of the proposed algorithm effectively sample the target conditional distribution with minimal impact on the quality of the samples compared to state-of-the-art methods. Additionally, the procedure can be used as a metric for the diversity of samples generated by a conditional GAN (cGAN) conditioned with continuous variables.

Monte-Carlo Tree Search (MCTS) has been applied successfully in many domains, including games. However, its performance is not uniform on all domains, and it also depends on how parameters that control the search are set. Parameter values that are optimal for a task might be sub-optimal for another. In a domain that tackles many games with different characteristics, like general game playing (GGP), selecting appropriate parameter settings is not a trivial task. Games are unknown to the player, thus, finding optimal parameters for a given game in advance is not feasible. Previous work has looked into tuning parameter values online, while the game is being played, showing some promising results. This tuning approach looks for optimal parameter values, balancing exploitation of values that performed well so far in the search and exploration of less sampled values. Continuously changing parameter values while performing the search, combined also with exploration of multiple values, introduces some randomization in the process. In addition, previous research indicates that adding randomization to certain components of MCTS might increase the diversification of the search and improve the performance. Therefore, this article investigates the effect of randomly selecting values for MCTS search-control parameters online among predefined sets of reasonable values. For the GGP domain, this article evaluates four different online parameter randomization strategies by comparing them with other methods to set parameter values: online parameter tuning, offline parameter tuning and sub-optimal parameter choices. Results on a set of 14 heterogeneous abstract games show that randomizing parameter values before each simulation has a positive effect on the search in some of the tested games, with respect to using fixed offline-tuned parameters. Moreover, results show a clear distinction between games for which online parameter tuning works best and games for which online randomization works best. In addition, the overall performance of online parameter randomization is closer to the one of online parameter turning than the one of sub-optimal parameter values, showing that online randomization is a reasonable parameter selection strategy. When analyzing the structure of the search trees generated by agents that use the different parameters selection strategies, it is clear that randomization causes MCTS to become more explorative, which is helpful for alignment games that present many winning paths in their trees. Online parameter tuning, instead, seems more suitable for games that present narrow winning paths and many losing paths.

Model predictive control (MPC) is increasingly being considered for control of fast systems and embedded applications. However, the MPC has some significant challenges for such systems. Its high computational complexity results in high power consumption from the control algorithm, which could account for a significant share of the energy resources in battery-powered embedded systems. The MPC parameters must be tuned, which is largely a trial-and-error process that affects the control performance, the robustness and the computational complexity of the controller to a high degree. In this paper, we propose a novel framework in which any parameter of the control algorithm can be jointly tuned using reinforcement learning(RL), with the goal of simultaneously optimizing the control performance and the power usage of the control algorithm. We propose the novel idea of optimizing the meta-parameters of MPCwith RL, i.e. parameters affecting the structure of the MPCproblem as opposed to the solution to a given problem. Our control algorithm is based on an event-triggered MPC where we learn when the MPC should be re-computed, and a dual mode MPC and linear state feedback control law applied in between MPC computations. We formulate a novel mixture-distribution policy and show that with joint optimization we achieve improvements that do not present themselves when optimizing the same parameters in isolation. We demonstrate our framework on the inverted pendulum control task, reducing the total computation time of the control system by 36% while also improving the control performance by 18.4% over the best-performing MPC baseline.

We propose a novel deep learning based surrogate model for solving high-dimensional uncertainty quantification and uncertainty propagation problems. The proposed deep learning architecture is developed by integrating the well-known U-net architecture with the Gaussian Gated Linear Network (GGLN) and referred to as the Gated Linear Network induced U-net or GLU-net. The proposed GLU-net treats the uncertainty propagation problem as an image to image regression and hence, is extremely data efficient. Additionally, it also provides estimates of the predictive uncertainty. The network architecture of GLU-net is less complex with 44\% fewer parameters than the contemporary works. We illustrate the performance of the proposed GLU-net in solving the Darcy flow problem under uncertainty under the sparse data scenario. We consider the stochastic input dimensionality to be up to 4225. Benchmark results are generated using the vanilla Monte Carlo simulation. We observe the proposed GLU-net to be accurate and extremely efficient even when no information about the structure of the inputs is provided to the network. Case studies are performed by varying the training sample size and stochastic input dimensionality to illustrate the robustness of the proposed approach.

Individual device loads and energy consumption feedback is one of the important approaches for pursuing users to save energy in residences. This can help in identifying faulty devices and wasted energy by devices when left On unused. The main challenge is to identity and estimate the energy consumption of individual devices without intrusive sensors on each device. Non-intrusive load monitoring (NILM) or energy disaggregation, is a blind source separation problem which requires a system to estimate the electricity usage of individual appliances from the aggregated household energy consumption. In this paper, we propose a novel deep neural network-based approach for performing load disaggregation on low frequency power data obtained from residential households. We combine a series of one-dimensional Convolutional Neural Networks and Long Short Term Memory (1D CNN-LSTM) to extract features that can identify active appliances and retrieve their power consumption given the aggregated household power value. We used CNNs to extract features from main readings in a given time frame and then used those features to classify if a given appliance is active at that time period or not. Following that, the extracted features are used to model a generation problem using LSTM. We train the LSTM to generate the disaggregated energy consumption of a particular appliance. Our neural network is capable of generating detailed feedback of demand-side, providing vital insights to the end-user about their electricity consumption. The algorithm was designed for low power offline devices such as ESP32. Empirical calculations show that our model outperforms the state-of-the-art on the Reference Energy Disaggregation Dataset (REDD).