In the U.S., corn is the most produced crop and has been an essential part of the American diet. To meet the demand for supply chain management and regional food security, accurate and timely large-scale corn yield prediction is attracting more attention in precision agriculture. Recently, remote sensing technology and machine learning methods have been widely explored for crop yield prediction. Currently, most county-level yield prediction models use county-level mean variables for prediction, ignoring much detailed information. Moreover, inconsistent spatial resolution between crop area and satellite sensors results in mixed pixels, which may decrease the prediction accuracy. Only a few works have addressed the mixed pixels problem in large-scale crop yield prediction. To address the information loss and mixed pixels problem, we developed a variational autoencoder (VAE) based multiple instance regression (MIR) model for large-scaled corn yield prediction. We use all unlabeled data to train a VAE and the well-trained VAE for anomaly detection. As a preprocess method, anomaly detection can help MIR find a better representation of every bag than traditional MIR methods, thus better performing in large-scale corn yield prediction. Our experiments showed that variational autoencoder based multiple instance regression (VAEMIR) outperformed all baseline methods in large-scale corn yield prediction. Though a suitable meta parameter is required, VAEMIR shows excellent potential in feature learning and extraction for large-scale corn yield prediction.

Numerical methods such as Finite element method [Hug12], Isogeomteric analysis [CHB09], and mesh-free methods [LJZ95, BLG94] are few of the conventional approaches employed in solving the Partial Differential Equations (PDEs) involved in computational solid mechanics problems. However, the ever-increasing sophistication of material models by incorporating more complex physics, such as modeling size-effect [FMAH94, AA20b] or dislocation density evolution [AZA20, Aro19, AA20a, AAA22, JABG20], or advanced materials such as composites and multicomponent alloys with spatially-varying material properties (heterogeneity) and direction dependent behavior (anisotropy) is bringing these numerical solvers to their limits. Hence, it is becoming a formidable task to perform simulations that can resolve the complex physical phenomena occurring at small spatial and temporal scales and accurately predict the macro-scale behavior of materials. Therefore, a cost-effective physicsbased surrogate model that allows the researchers to perform simulations on a coarse mesh without sacrificing accuracy will be highly beneficial for many reasons. First, researchers can choose to run their simulations at a lower resolution (online stage) and later reconstruct the solution back to the target resolution (offline stage). This will significantly reduce the computational expense during the online stage, thus accelerating the process of scientific investigation and discovery. Second, the surrogate model based on data super-resolution can also be used to enhance outputs from experimental techniques for full-field displacement and strain measurement such as Digital Image Correlation (DIC) which would allow researchers to generate and store a small fraction of data. Recent advances in Deep Learning (DL) and Physics-Informed Neural Networks (PINN) [RPK17, RPK19] make it a promising tool to tackle this computational challenge.

Machine learning has become a promising approach for molecular modeling. Positional quantities, such as interatomic distances and bond angles, play a crucial role in molecule physics. The existing works rely on careful manual design of their representation. To model the complex nonlinearity in predicting molecular properties in an more end-to-end approach, we propose to encode the positional quantities with a learnable embedding that is continuous and differentiable. A regularization technique is employed to encourage embedding smoothness along the physical dimension. We experiment with a variety of molecular property and force field prediction tasks. Improved performance is observed for three different model architectures after plugging in the proposed positional encoding method. In addition, the learned positional encoding allows easier physics-based interpretation. We observe that tasks of similar physics have the similar learned positional encoding.

Temporal sequences of satellite images constitute a highly valuable and abundant resource to analyze a given region. However, the labeled data needed to train most machine learning models are scarce and difficult to obtain. In this context, the current work investigates a fully unsupervised methodology that, given a sequence of images, learns a semantic embedding and then, creates a partition of the ground according to its semantic properties and its evolution over time. We illustrate the methodology by conducting the semantic analysis of a sequence of satellite images of a region of Navarre (Spain). The proposed approach reveals a novel broad perspective of the land, where potentially large areas that share both a similar semantic and a similar temporal evolution are connected in a compact and well-structured manner. The results also show a close relationship between the allocation of the clusters in the geographic space and their allocation in the embedded spaces. The semantic analysis is completed by obtaining the representative sequence of tiles corresponding to each cluster, the linear interpolation between related areas, and a graph that shows the relationships between the clusters, providing a concise semantic summary of the whole region.

Digital-industrial innovation can now prove its mettle. For the last ten years, digital-industrial innovation has been seen as the poor, unglamorous cousin of pure digital innovation. It arrived late to the party. Its applications were developed on factory floors, in what was looked down upon as "old" economy. General Electric GE, which pioneered what it called the Industrial Internet, found that attracting software talent was one of its biggest difficulties.

This proposed "equitism" sounds aspirational. The transportation pods, zero-landfill, and sustainable-energy plans all seem technologically feasible. I worry that finding humans today with the altruism, will, and ability to administer the endowment Lore envisions will be more difficult than achieving those desert food-production and water-use goals. While we understand the appeal of a clean-slate approach to experiment, surely the wealthy can find ways of investing and contributing in our broader society that isn't holed off from the rest of the world. But at least Telosa sounds way nicer than The Line.

Quantum Progress 7. Progress in Green Technology 8. Robots Will Become More Human 9. Progress in Autonomous System 10. What are the biggest tech trends you need to be ready for in 2023? Learn what technologies will gain the most traction in the new year and what you need to do to prepare. Stay updated on last news about Artificial Intelligence. Check your inbox or spam folder to confirm your subscription.

The spatiotemporal dynamics of turbulent flows is chaotic and difficult to predict. This makes the design of accurate and stable reduced-order models challenging. The overarching objective of this paper is to propose a nonlinear decomposition of the turbulent state for a reduced-order representation of the dynamics. We divide the turbulent flow into a spatial problem and a temporal problem. First, we compute the latent space, which is the manifold onto which the turbulent dynamics live (i.e., it is a numerical approximation of the turbulent attractor). The latent space is found by a series of nonlinear filtering operations, which are performed by a convolutional autoencoder (CAE). The CAE provides the decomposition in space. Second, we predict the time evolution of the turbulent state in the latent space, which is performed by an echo state network (ESN). The ESN provides the decomposition in time. Third, by assembling the CAE and the ESN, we obtain an autonomous dynamical system: the convolutional autoncoder echo state network (CAE-ESN). This is the reduced-order model of the turbulent flow. We test the CAE-ESN on a two-dimensional flow. We show that, after training, the CAE-ESN (i) finds a latent-space representation of the turbulent flow that has less than 1% of the degrees of freedom than the physical space; (ii) time-accurately and statistically predicts the flow in both quasiperiodic and turbulent regimes; (iii) is robust for different flow regimes (Reynolds numbers); and (iv) takes less than 1% of computational time to predict the turbulent flow than solving the governing equations. This work opens up new possibilities for nonlinear decompositions and reduced-order modelling of turbulent flows from data.

We focus on robot navigation in crowded environments. The challenge of predicting the motion of a crowd around a robot makes it hard to ensure human safety and comfort. Recent approaches often employ end-to-end techniques for robot control or deep architectures for high-fidelity human motion prediction. While these methods achieve important performance benchmarks in simulated domains, dataset limitations and high sample complexity tend to prevent them from transferring to real-world environments. Our key insight is that a low-dimensional representation that captures critical features of crowd-robot dynamics could be sufficient to enable a robot to wind through a crowd smoothly. To this end, we mathematically formalize the act of passing between two agents as a rotation, using a notion of topological invariance. Based on this formalism, we design a cost functional that favors robot trajectories contributing higher passing progress and penalizes switching between different sides of a human. We incorporate this functional into a model predictive controller that employs a simple constant-velocity model of human motion prediction. This results in robot motion that accomplishes statistically significantly higher clearances from the crowd compared to state-of-the-art baselines while maintaining competitive levels of efficiency, across extensive simulations and challenging real-world experiments on a self-balancing robot.

This paper studies how to train machine-learning models that directly approximate the optimal solutions of constrained optimization problems. This is an empirical risk minimization under constraints, which is challenging as training must balance optimality and feasibility conditions. Supervised learning methods often approach this challenge by training the model on a large collection of pre-solved instances. This paper takes a different route and proposes the idea of Primal-Dual Learning (PDL), a self-supervised training method that does not require a set of pre-solved instances or an optimization solver for training and inference. Instead, PDL mimics the trajectory of an Augmented Lagrangian Method (ALM) and jointly trains primal and dual neural networks. Being a primal-dual method, PDL uses instance-specific penalties of the constraint terms in the loss function used to train the primal network. Experiments show that, on a set of nonlinear optimization benchmarks, PDL typically exhibits negligible constraint violations and minor optimality gaps, and is remarkably close to the ALM optimization. PDL also demonstrated improved or similar performance in terms of the optimality gaps, constraint violations, and training times compared to existing approaches.