Inverse deep learning methods and benchmarks for artificial electromagnetic material design.

In this work we investigate the use of deep inverse models (DIMs) for designing artificial electromagnetic materials (AEMs) - such as metamaterials, photonic crystals, and plasmonics - to achieve some desired scattering properties (e.g., transmission or reflection spectrum). DIMs are deep neural networks (i.e., deep learning models) that are specially-designed to solve ill-posed inverse problems. There has recently been tremendous growth in the use of DIMs for solving AEM design problems however there has been little comparison of these approaches to examine their absolute and relative performance capabilities. In this work we compare eight state-of-the-art DIMs on three unique AEM design problems, including two models that are novel to the AEM community. Our results indicate that DIMs can rapidly produce accurate designs to achieve a custom desired scattering on all three problems. Although no single model always performs best, the Neural-Adjoint approach achieves the best overall performance across all problem settings. As a final contribution we show that not all AEM design problems are ill-posed, and in such cases a conventional deep neural network can perform better than DIMs. We recommend that a deep neural network is always employed as a simple baseline approach when addressing AEM design problems. We publish python code for our AEM simulators and our DIMs to enable easy replication of our results, and benchmarking of new DIMs by the AEM community.

Neural-adjoint method for the inverse design of all-dielectric metasurfaces.

All-dielectric metasurfaces exhibit exotic electromagnetic responses, similar to those obtained with metal-based metamaterials. Research in all-dielectric metasurfaces currently uses relatively simple unit-cell designs, but increased geometrical complexity may yield even greater scattering states. Although machine learning has recently been applied to the design of metasurfaces with impressive results, the much more challenging task of finding a geometry that yields a desired spectra remains largely unsolved. We propose and demonstrate a method capable of finding accurate solutions to ill-posed inverse problems, where the conditions of existence and uniqueness are violated. A specific example of finding the metasurface geometry which yields a radiant exitance matching the external quantum efficiency of gallium antimonide is demonstrated. We also show how the neural-adjoint method can intelligently grow the design search space to include designs that increasingly and accurately approximate the desired scattering response. The neural-adjoint method is not restricted to the case demonstrated and may be applied to plasmonics, photonic crystal, and other artificial electromagnetic materials.

Deep learning for accelerated all-dielectric metasurface design.

Deep learning has risen to the forefront of many fields in recent years, overcoming challenges previously considered intractable with conventional means. Materials discovery and optimization is one such field, but significant challenges remain, including the requirement of large labeled datasets and one-to-many mapping that arises in solving the inverse problem. Here we demonstrate modeling of complex all-dielectric metasurface systems with deep neural networks, using both the metasurface geometry and knowledge of the underlying physics as inputs. Our deep learning network is highly accurate, achieving an average mean square error of only 1.16 × 10-3 and is over five orders of magnitude faster than conventional electromagnetic simulation software. We further develop a novel method to solve the inverse modeling problem, termed fast forward dictionary search (FFDS), which offers tremendous controls to the designer and only requires an accurate forward neural network model. These techniques significantly increase the viability of more complex all-dielectric metasurface designs and provide opportunities for the future of tailored light matter interactions.

Distributed solar photovoltaic array location and extent dataset for remote sensing object identification.

Earth-observing remote sensing data, including aerial photography and satellite imagery, offer a snapshot of the world from which we can learn about the state of natural resources and the built environment. The components of energy systems that are visible from above can be automatically assessed with these remote sensing data when processed with machine learning methods. Here, we focus on the information gap in distributed solar photovoltaic (PV) arrays, of which there is limited public data on solar PV deployments at small geographic scales. We created a dataset of solar PV arrays to initiate and develop the process of automatically identifying solar PV locations using remote sensing imagery. This dataset contains the geospatial coordinates and border vertices for over 19,000 solar panels across 601 high-resolution images from four cities in California. Dataset applications include training object detection and other machine learning algorithms that use remote sensing imagery, developing specific algorithms for predictive detection of distributed PV systems, estimating installed PV capacity, and analysis of the socioeconomic correlates of PV deployment.

The effect of class imbalance on case selection for case-based classifiers: an empirical study in the context of medical decision support.

Case selection is a useful approach for increasing the efficiency and performance of case-based classifiers. Multiple techniques have been designed to perform case selection. This paper empirically investigates how class imbalance in the available set of training cases can impact the performance of the resulting classifier as well as properties of the selected set. In this study, the experiments are performed using a dataset for the problem of detecting breast masses in screening mammograms. The classification problem was binary and we used a k-nearest neighbor classifier. The classifier's performance was evaluated using the receiver operating characteristic (ROC) area under the curve (AUC) measure. The experimental results indicate that although class imbalance reduces the performance of the derived classifier and the effectiveness of selection at improving overall classifier performance, case selection can still be beneficial, regardless of the level of class imbalance.

Comparative analysis of instance selection algorithms for instance-based classifiers in the context of medical decision support.

When constructing a pattern classifier, it is important to make best use of the instances (a.k.a. cases, examples, patterns or prototypes) available for its development. In this paper we present an extensive comparative analysis of algorithms that, given a pool of previously acquired instances, attempt to select those that will be the most effective to construct an instance-based classifier in terms of classification performance, time efficiency and storage requirements. We evaluate seven previously proposed instance selection algorithms and compare their performance to simple random selection of instances. We perform the evaluation using k-nearest neighbor classifier and three classification problems: one with simulated Gaussian data and two based on clinical databases for breast cancer detection and diagnosis, respectively. Finally, we evaluate the impact of the number of instances available for selection on the performance of the selection algorithms and conduct initial analysis of the selected instances. The experiments show that for all investigated classification problems, it was possible to reduce the size of the original development dataset to less than 3% of its initial size while maintaining or improving the classification performance. Random mutation hill climbing emerges as the superior selection algorithm. Furthermore, we show that some previously proposed algorithms perform worse than random selection. Regarding the impact of the number of instances available for the classifier development on the performance of the selection algorithms, we confirm that the selection algorithms are generally more effective as the pool of available instances increases. In conclusion, instance selection is generally beneficial for instance-based classifiers as it can improve their performance, reduce their storage requirements and improve their response time. However, choosing the right selection algorithm is crucial.