Showing papers by "Toyota published in 2018"

Accelerating the discovery of materials for clean energy in the era of smart automation

Abstract: The discovery and development of novel materials in the field of energy are essential to accelerate the transition to a low-carbon economy. Bringing recent technological innovations in automation, robotics and computer science together with current approaches in chemistry, materials synthesis and characterization will act as a catalyst for revolutionizing traditional research and development in both industry and academia. This Perspective provides a vision for an integrated artificial intelligence approach towards autonomous materials discovery, which, in our opinion, will emerge within the next 5 to 10 years. The approach we discuss requires the integration of the following tools, which have already seen substantial development to date: high-throughput virtual screening, automated synthesis planning, automated laboratories and machine learning algorithms. In addition to reducing the time to deployment of new materials by an order of magnitude, this integrated approach is expected to lower the cost associated with the initial discovery. Thus, the price of the final products (for example, solar panels, batteries and electric vehicles) will also decrease. This in turn will enable industries and governments to meet more ambitious targets in terms of reducing greenhouse gas emissions at a faster pace. The discovery and development of advanced materials are imperative for the clean energy sector. We envision that a closed-loop approach, which combines high-throughput computation, artificial intelligence and advanced robotics, will sizeably reduce the time to deployment and the costs associated with materials development.

A strategy to apply machine learning to small datasets in materials science

Abstract: There is growing interest in applying machine learning techniques in the research of materials science. However, although it is recognized that materials datasets are typically smaller and sometimes more diverse compared to other fields, the influence of availability of materials data on training machine learning models has not yet been studied, which prevents the possibility to establish accurate predictive rules using small materials datasets. Here we analyzed the fundamental interplay between the availability of materials data and the predictive capability of machine learning models. Instead of affecting the model precision directly, the effect of data size is mediated by the degree of freedom (DoF) of model, resulting in the phenomenon of association between precision and DoF. The appearance of precision–DoF association signals the issue of underfitting and is characterized by large bias of prediction, which consequently restricts the accurate prediction in unknown domains. We proposed to incorporate the crude estimation of property in the feature space to establish ML models using small sized materials data, which increases the accuracy of prediction without the cost of higher DoF. In three case studies of predicting the band gap of binary semiconductors, lattice thermal conductivity, and elastic properties of zeolites, the integration of crude estimation effectively boosted the predictive capability of machine learning models to state-of-art levels, demonstrating the generality of the proposed strategy to construct accurate machine learning models using small materials dataset. Machine learning can be useful for materials prediction if crude estimations of the outcome are integrated in the code. Machine learning has been attracting tremendous attention lately due to its predictive power; evidence suggests it is directly proportional to the size of the available datasets. Machine learning can be useful in predicting new materials and novel properties, but materials sets tend to be smaller and more diverse than other fields. Ying Zhang and Chen Ling from the Toyota Research Institute of North America report that these small datasets affect the freedom of the algorithms and thus limit their predictive capabilities. In order to counterbalance the effect, they suggest introducing in the code crude estimations of the targeted property, obtained by other means.

BOP: Benchmark for 6D Object Pose Estimation

Abstract: We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image. The training data consists of a texture-mapped 3D object model or images of the object in known 6D poses. The benchmark comprises of: i) eight datasets in a unified format that cover different practical scenarios, including two new datasets focusing on varying lighting conditions, ii) an evaluation methodology with a pose-error function that deals with pose ambiguities, iii) a comprehensive evaluation of 15 diverse recent methods that captures the status quo of the field, and iv) an online evaluation system that is open for continuous submission of new results. The evaluation shows that methods based on point-pair features currently perform best, outperforming template matching methods, learning-based methods and methods based on 3D local features. The project website is available at bop.felk.cvut.cz.

BOP: Benchmark for 6D Object Pose Estimation

Abstract: We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image. The training data consists of a texture-mapped 3D object model or images of the object in known 6D poses. The benchmark comprises of: (i) eight datasets in a unified format that cover different practical scenarios, including two new datasets focusing on varying lighting conditions, (ii) an evaluation methodology with a pose-error function that deals with pose ambiguities, (iii) a comprehensive evaluation of 15 diverse recent methods that captures the status quo of the field, and (iv) an online evaluation system that is open for continuous submission of new results. The evaluation shows that methods based on point-pair features currently perform best, outperforming template matching methods, learning-based methods and methods based on 3D local features. The project website is available at bop.felk.cvut.cz.

Simulation-based Adversarial Test Generation for Autonomous Vehicles with Machine Learning Components

Abstract: Many organizations are developing autonomous driving systems, which are expected to be deployed at a large scale in the near future. Despite this, there is a lack of agreement on appropriate methods to test, debug, and certify the performance of these systems. One of the main challenges is that many autonomous driving systems have machine learning (ML) components, such as deep neural networks, for which formal properties are difficult to characterize. We present a testing framework that is compatible with test case generation and automatic falsification methods, which are used to evaluate cyber-physical systems. We demonstrate how the framework can be used to evaluate closed-loop properties of an autonomous driving system model that includes the ML components, all within a virtual environment. We demonstrate how to use test case generation methods, such as covering arrays, as well as requirement falsification methods to automatically identify problematic test scenarios. The resulting framework can be used to increase the reliability of autonomous driving systems.

Street-view change detection with deconvolutional networks

Abstract: We propose a system for performing structural change detection in street-view videos captured by a vehicle-mounted monocular camera over time. Our approach is motivated by the need for more frequent and efficient updates in the large-scale maps used in autonomous vehicle navigation. Our method chains a multi-sensor fusion SLAM and fast dense 3D reconstruction pipeline, which provide coarsely registered image pairs to a deep Deconvolutional Network (DN) for pixel-wise change detection. We investigate two DN architectures for change detection, the first one is based on the idea of stacking contraction and expansion blocks while the second one is based on the idea of Fully Convolutional Networks. To train and evaluate our networks we introduce a new urban change detection dataset which is an order of magnitude larger than existing datasets and contains challenging changes due to seasonal and lighting variations. Our method outperforms existing literature on this dataset, which we make available to the community, and an existing panoramic change detection dataset, demonstrating its wide applicability.

All-Solid-State Batteries with Thick Electrode Configurations

Abstract: We report the preparation of thick electrode all-solid-state lithium-ion cells in which a large geometric capacity of 15.7 mAh cm–2 was achieved at room temperature using a 600 μm-thick cathode layer. The effect of ionic conductivity on the discharge performance was then examined using two different materials for the solid electrolyte. Furthermore, important morphological information regarding the tortuosity factor was electrochemically extracted from the capacity-current data. The effect of tortuosity on cell performance was also quantitatively discussed.

Deep Model-Based 6D Pose Refinement in RGB

Abstract: We present a novel approach for model-based 6D pose refinement in color data. Building on the established idea of contour-based pose tracking, we teach a deep neural network to predict a translational and rotational update. At the core, we propose a new visual loss that drives the pose update by aligning object contours, thus avoiding the definition of any explicit appearance model. In contrast to previous work our method is correspondence-free, segmentation-free, can handle occlusion and is agnostic to geometrical symmetry as well as visual ambiguities. Additionally, we observe a strong robustness towards rough initialization. The approach can run in real-time and produces pose accuracies that come close to 3D ICP without the need for depth data. Furthermore, our networks are trained from purely synthetic data and will be published together with the refinement code at http://campar.in.tum.de/Main/FabianManhardt to ensure reproducibility.

Inverse Multipath Fingerprinting for Millimeter Wave V2I Beam Alignment

Abstract: Efficient beam alignment is a crucial component in millimeter wave systems with analog beamforming, especially in fast-changing vehicular settings. This paper proposes to use the vehicle's position (e.g., available via GPS) to query a multipath fingerprint database, which provides prior knowledge of potential pointing directions for reliable beam alignment. The approach is the inverse of fingerprinting localization, where the measured multipath signature is compared to the fingerprint database to retrieve the most likely position. The power loss probability is introduced as a metric to quantify misalignment accuracy and is used for optimizing candidate beam selection. Two candidate beam selection methods are developed, where one is a heuristic while the other minimizes the misalignment probability. The proposed beam alignment is evaluated using realistic channels generated from a commercial ray-tracing simulator. Using the generated channels, an extensive investigation is provided, which includes the required measurement sample size to build an effective fingerprint, the impact of measurement noise, the sensitivity to changes in traffic density, and beam alignment overhead comparison with IEEE 802.11ad as the baseline. Using the concept of beam coherence time, which is the duration between two consecutive beam alignments, and parameters of IEEE 802.11ad, the overhead is compared in the mobility context. The results show that while the proposed approach provides increasing rates with larger antenna arrays, IEEE 802.11ad has decreasing rates due to the higher beam training overhead that eats up a large portion of the beam coherence time, which becomes shorter with increasing mobility.

Benchmarking high surface area electrocatalysts in a gas diffusion electrode: measurement of oxygen reduction activities under realistic conditions

Abstract: In this work, we introduce the application of gas diffusion electrodes (GDE) for benchmarking the electrocatalytic performance of high surface area fuel cell catalysts. It is demonstrated that GDEs offer several inherent advantages over the state-of-the-art technique, i.e. thin film rotating disk electrode (TF-RDE) measurements for fast fuel cell catalyst evaluation. The most critical advantage is reactant mass transport. While in RDE measurements the reactant mass transport is severely limited by the gas solubility of the reactant in the electrolyte, GDEs enable reactant transport rates similar to technical fuel cell devices. Hence, in contrast to TF-RDE measurements, performance data obtained from GDE measurements can be directly compared to membrane electrode assembly (MEA) tests. Therefore, the application of GDEs for the testing of fuel cell catalysts closes the gap between catalyst research in academia and real applications.

Structural basis for cofilin binding and actin filament disassembly

Abstract: Actin depolymerizing factor (ADF) and cofilin accelerate actin dynamics by severing and disassembling actin filaments. Here, we present the 3.8 A resolution cryo-EM structure of cofilactin (cofilin-decorated actin filament). The actin subunit structure of cofilactin (C-form) is distinct from those of F-actin (F-form) and monomeric actin (G-form). During the transition between these three conformations, the inner domain of actin (subdomains 3 and 4) and the majority of subdomain 1 move as two separate rigid bodies. The cofilin–actin interface consists of three distinct parts. Based on the rigid body movements of actin and the three cofilin–actin interfaces, we propose models for the cooperative binding of cofilin to actin, preferential binding of cofilin to ADP-bound actin filaments and cofilin-mediated severing of actin filaments.

Discovery of superconductivity in quasicrystal

Abstract: Superconductivity is ubiquitous as evidenced by the observation in many crystals including carrier-doped oxides and diamond. Amorphous solids are no exception. However, it remains to be discovered in quasicrystals, in which atoms are ordered over long distances but not in a periodically repeating arrangement. Here we report electrical resistivity, magnetization, and specific-heat measurements of Al–Zn–Mg quasicrystal, presenting convincing evidence for the emergence of bulk superconductivity at a very low transition temperature of $$T_{\rm c} \cong 0.05$$ K. We also find superconductivity in its approximant crystals, structures that are periodic, but that are very similar to quasicrystals. These observations demonstrate that the effective interaction between electrons remains attractive under variation of the atomic arrangement from periodic to quasiperiodic one. The discovery of the superconducting quasicrystal, in which the fractal geometry interplays with superconductivity, opens the door to a new type of superconductivity, fractal superconductivity. Superconductivity is evidenced in crystals and amorphous solids, but remains to be discovered in quasicrystals. Here, Kamiya et al. report the emergence of bulk superconductivity in Al-Zn-Mg quasicrystal at a very low transition temperature about 0.05 K.

Learning to Zoom: a Saliency-Based Sampling Layer for Neural Networks

Abstract: We introduce a saliency-based distortion layer for convolutional neural networks that helps to improve the spatial sampling of input data for a given task Our differentiable layer can be added as a preprocessing block to existing task networks and trained altogether in an end-to-end fashion The effect of the layer is to efficiently estimate how to sample from the original data in order to boost task performance For example, for an image classification task in which the original data might range in size up to several megapixels, but where the desired input images to the task network are much smaller, our layer learns how best to sample from the underlying high resolution data in a manner which preserves task-relevant information better than uniform downsampling This has the effect of creating distorted, caricature-like intermediate images, in which idiosyncratic elements of the image that improve task performance are zoomed and exaggerated Unlike alternative approaches such as spatial transformer networks, our proposed layer is inspired by image saliency, computed efficiently from uniformly downsampled data, and degrades gracefully to a uniform sampling strategy under uncertainty We apply our layer to improve existing networks for the tasks of human gaze estimation and fine-grained object classification Code for our method is available in: http://githubcom/recasens/Saliency-Sampler

Dynamic Humanoid Locomotion: A Scalable Formulation for HZD Gait Optimization

Abstract: Hybrid zero dynamics (HZD) has emerged as a popular framework for dynamic walking but has significant implementation difficulties when applied to the high degrees of freedom humanoids. The primary impediment is the process of gait design—it is difficult for optimizers to converge on a viable set of virtual constraints defining a gait. This paper presents a methodology that allows for fast and reliable generation of dynamic robotic walking gaits through the HZD framework, even in the presence of underactuation. Specifically, we describe an optimization formulation that builds upon the novel combination of HZD and direct collocation methods. Furthermore, achieving a scalable implementation required developing a defect-variable substitution formulation to simplify expressions, which ultimately allows us to generate compact analytic Jacobians of the constraints. We experimentally validate our methodology on an underactuated humanoid, DURUS, a spring-legged machine designed to facilitate energy-economical walking. We show that the optimization approach, in concert with the HZD framework, yields dynamic and stable walking gaits in hardware with a total electrical cost of transport of 1.33.

Traffic Signal Control Based on Reinforcement Learning with Graph Convolutional Neural Nets

Abstract: Traffic signal control can mitigate traffic congestion and reduce travel time. A model-free reinforcement learning (RL) approach is a powerful framework for learning a responsive traffic control policy for short-term traffic demand changes without prior environmental knowledge. Previous RL approaches could handle high-dimensional feature space using a standard neural network, e.g., a convolutional neural network; however, to control traffic on a road network with multiple intersections, the geometric features between roads had to be created manually. Rather than using manually crafted geometric features, we developed an RL-based traffic signal control method that employs a graph convolutional neural network (GCNN). GCNNs can automatically extract features considering the traffic features between distant roads by stacking multiple neural network layers. We numerically evaluated the proposed method in a six-intersection environment. The results demonstrate that the proposed method can find comparable policies twice as fast as the conventional RL method with a neural network and can adapt to more extensive traffic demand changes.

Effect of the Side-Chain Structure of Perfluoro-Sulfonic Acid Ionomers on the Oxygen Reduction Reaction on the Surface of Pt

Abstract: The effect of side-chain structures in perfluoro-sulfonic acid ionomers on the adsorption of the terminal sulfonate moiety on the surface of Pt is investigated with voltammetry and surface-enhanced infrared absorption spectroscopy (SEIRAS). Analyses with low-molecular-weight model anions with and without an ether group in the perfluoro-alkyl chain indicate that the anions are adsorbed on Pt through one or two oxygen atom(s) of the terminal sulfonate group and that the oxygen atom of the ether group also interacts with the Pt surface, leading to stronger adsorption of the anions with an ether group. On the basis of the results obtained with the model anions, the adsorption of the terminal sulfonate moieties in perfluorinated sulfonic acid ionomers and its effect on oxygen reduction reaction (ORR) is discussed. It is shown that the ionomers having longer side chains more strongly block ORR due to the flexibility of the side chains.

Distributed Conflict Resolution for Connected Autonomous Vehicles

Abstract: This paper proposes a novel communication-enabled distributed conflict resolution mechanism in order for a group of connected autonomous vehicles to navigate safely and efficiently in intersections without any traffic manager. The conflict resolution strategy for individual vehicle is decoupled temporally. In a decision maker, the vehicle computes the desired time slots to pass the conflict zones by solving a conflict graph locally based on the broadcasted information from other vehicles. In a motion planner, the vehicle computes the desired speed profile by solving a temporal optimization problem constrained in the desired time slot. The estimated time to occupy the conflict zones given the new speed profile is then broadcasted again. It is proved that the aggregation of these local decisions solves the conflicts globally. Theoretically, this method provides an efficient parallel mechanism to obtain local optimal solutions of a large-scale optimization problem (e.g., multivehicle navigation). Application-wise, as demonstrated by extensive simulation, this mechanism increases the efficiency of autonomous vehicles in terms of smaller delay time, as well as the efficiency of the traffic in terms of larger throughput when there is no traffic manager to mediate the conflicts.

Solar-Driven Photocatalytic CO2 Reduction in Water Utilizing a Ruthenium Complex Catalyst on p-Type Fe2O3 with a Multiheterojunction

Abstract: A hybrid photocathode that consists of a ruthenium complex catalyst and a p-type semiconductor composed of earth-abundant elements, N,Zn-codoped Fe2O3, with a multiheterojunction structure (TiO2/N,Zn-Fe2O3/Cr2O3) was developed for the reduction of CO2 in aqueous solution with application of an electrical bias under simulated solar light irradiation. The TiO2 layer prevents contact between N,Zn-Fe2O3 and the electrolyte, so that dissolution of N,Zn-Fe2O3 by photoelectrochemical (PEC) self-reduction cannot occur. Both TiO2 and Cr2O3 significantly enhanced the cathodic photocurrent by tuning the band bending in N,Zn-Fe2O3. The use of a Ru complex with an electronic network provided by polypyrrole improved the performance and resulted in a stable photocurrent of 150 μA cm–2 for the production of HCOOH, CO, and a small amount of H2 under 1 sun irradiation with application of 0.1 V vs the reversible hydrogen electrode (RHE). The total amount of generated HCOOH, CO, and H2, two-electron-reduction products, was e...

Electrochemical Intercalation of Calcium and Magnesium in TiS2: Fundamental Studies Related to Multivalent Battery Applications

Abstract: A comparative study of the electrochemical intercalation of Ca2+ and Mg2+ in layered TiS2 using alkylcarbonate-based electrolytes is reported, and for the first time, reversible electrochemical Ca2+ insertion is proved in this compound using both X-ray diffraction and differential absorption X-ray tomography at the Ca L2 edge. Different new phases are formed upon M2+ insertion that are structurally characterized, their amount and composition being dependent on M2+ and the experimental conditions. The first phase formed upon reduction is found to be the result of an ion-solvated intercalation mechanism, with solvent molecule(s) being cointercalated with the M2+ cation. Upon further reduction, new non-cointercalated calcium-containing phases seem to form at the expense of unreacted TiS2. The calculated activation energy barriers for Ca2+ migration in TiS2 (0.75 eV) are lower than those previously reported for Mg (1.14 eV) at the dilute limit and within the CdI2 structural type. DFT results indicate that the...

Area-selective atomic layer deposition of metal oxides on noble metals through catalytic oxygen activation

Abstract: Area-selective atomic layer deposition (ALD) is envisioned to play a key role in next-generation semiconductor processing and can also provide new opportunities in the field of catalysis. In this work, we developed an approach for the area-selective deposition of metal oxides on noble metals. Using O2 gas as co-reactant, area-selective ALD has been achieved by relying on the catalytic dissociation of the oxygen molecules on the noble metal surface, while no deposition takes place on inert surfaces that do not dissociate oxygen (i.e., SiO2, Al2O3, Au). The process is demonstrated for selective deposition of iron oxide and nickel oxide on platinum and iridium substrates. Characterization by in situ spectroscopic ellipsometry, transmission electron microscopy, scanning Auger electron spectroscopy, and X-ray photoelectron spectroscopy confirms a very high degree of selectivity, with a constant ALD growth rate on the catalytic metal substrates and no deposition on inert substrates, even after 300 ALD cycles. We demonstrate the area-selective ALD approach on planar and patterned substrates and use it to prepare Pt/Fe2O3 core/shell nanoparticles. Finally, the approach is proposed to be extendable beyond the materials presented here, specifically to other metal oxide ALD processes for which the precursor requires a strong oxidizing agent for growth.

Variational Autoencoder for End-to-End Control of Autonomous Driving with Novelty Detection and Training De-biasing

Abstract: This paper introduces a new method for end-to-end training of deep neural networks (DNNs) and evaluates it in the context of autonomous driving. DNN training has been shown to result in high accuracy for perception to action learning given sufficient training data. However, the trained models may fail without warning in situations with insufficient or biased training data. In this paper, we propose and evaluate a novel architecture for self-supervised learning of latent variables to detect the insufficiently trained situations. Our method also addresses training data imbalance, by learning a set of underlying latent variables that characterize the training data and evaluate potential biases. We show how these latent distributions can be leveraged to adapt and accelerate the training pipeline by training on only a fraction of the total dataset. We evaluate our approach on a challenging dataset for driving. The data is collected from a full-scale autonomous vehicle. Our method provides qualitative explanation for the latent variables learned in the model. Finally, we show how our model can be additionally trained as an end-to-end controller, directly outputting a steering control command for an autonomous vehicle.

Significant Enhancement of Near-Field Electromagnetic Heat Transfer in a Multilayer Structure through Multiple Surface-States Coupling.

Abstract: We show that near-field electromagnetic heat transfer between multilayer thermal bodies can be significantly enhanced by the contributions of surface states at multiple surfaces. As a demonstration, we show that when one of the materials forming the multilayer structure is described by the Drude model, and the other one is a vacuum, at the same gap spacing the resulting heat transfer can be up to 40 times higher as compared to that between two semi-infinite materials described by the same Drude model. Moreover, this system can exhibit a nonmonotonic dependency in its heat transfer coefficient as a function of the middle gap spacing. The enhancement effect in the system persists for realistic materials.

SuperDepth: Self-Supervised, Super-Resolved Monocular Depth Estimation

Abstract: Recent techniques in self-supervised monocular depth estimation are approaching the performance of supervised methods, but operate in low resolution only. We show that high resolution is key towards high-fidelity self-supervised monocular depth prediction. Inspired by recent deep learning methods for Single-Image Super-Resolution, we propose a sub-pixel convolutional layer extension for depth super-resolution that accurately synthesizes high-resolution disparities from their corresponding low-resolution convolutional features. In addition, we introduce a differentiable flip-augmentation layer that accurately fuses predictions from the image and its horizontally flipped version, reducing the effect of left and right shadow regions generated in the disparity map due to occlusions. Both contributions provide significant performance gains over the state-of-the-art in self-supervised depth and pose estimation on the public KITTI benchmark. A video of our approach can be found at this https URL.

Density Functional Theory (DFT) and Experimental Evidences of Metal–Support Interaction in Platinum Nanoparticles Supported on Nitrogen- and Sulfur-Doped Mesoporous Carbons: Synthesis, Activity, and Stability

Abstract: In this paper, we report a comprehensive investigation of Pt nanoparticles (NPs) deposition on nitrogen- and sulfur-doped or codoped mesoporous carbons (N-MC, S-MC, and N,S-MC) to develop active and durable oxygen reduction catalysts for fuel cells. N-MC, S-MC, and N,S-MC were prepared by employing mesoporous silica as hard template and suitable organic precursors. Pt NPs were deposited by solid-state reduction of platinum acetylacetonate under N2/H2 flow on the three different supports. Pt NPs resulted to be well-dispersed over the doped MC supports with size distributions (from 1.8 nm to 3.5 nm) that are dependent on the type of doping heteroatom (N, S, or N and S). The influence of nitrogen and/or sulfur incorporated into the carbon matrix on the nucleation and growth of Pt NPs was also rationalized based on density functional theory (DFT) simulations. They highlighted that both nitrogen and sulfur increase the interactions between Pt and carbon support, but the interaction decreases as the nitrogen an...

Bayesian-Driven First-Principles Calculations for Accelerating Exploration of Fast Ion Conductors for Rechargeable Battery Application

Abstract: Safe and robust batteries are urgently requested today for power sources of electric vehicles. Thus, a growing interest has been noted for fabricating those with solid electrolytes. Materials search by density functional theory (DFT) methods offers great promise for finding new solid electrolytes but the evaluation is known to be computationally expensive, particularly on ion migration property. In this work, we proposed a Bayesian-optimization-driven DFT-based approach to efficiently screen for compounds with low ion migration energies ( $${{\boldsymbol{E}}}_{{\boldsymbol{b}}}{\boldsymbol{)}}$$ . We demonstrated this on 318 tavorite-type Li- and Na-containing compounds. We found that the scheme only requires ~30% of the total DFT- $${{\boldsymbol{E}}}_{{\boldsymbol{b}}}$$ evaluations on the average to recover the optimal compound ~90% of the time. Its recovery performance for desired compounds in the tavorite search space is ~2× more than random search (i.e., for $${{\boldsymbol{E}}}_{{\boldsymbol{b}}}$$ < 0.3 eV). Our approach offers a promising way for addressing computational bottlenecks in large-scale material screening for fast ionic conductors.

Low-Energy Electrocatalytic CO2 Reduction in Water over Mn-Complex Catalyst Electrode Aided by a Nanocarbon Support and K+ Cations

Abstract: Electrocatalytic CO2 reduction over a Mn-complex catalyst in an aqueous solution was achieved at very low energy with a combination of multiwalled carbon nanotubes (MWCNTs) and K+ cations. Although the bare Mn-complex did not function as a catalyst in an aqueous solution, the combined Mn-complex/MWCNT cathode promoted electrocatalytic CO2 reduction at an overpotential of 100 mV where neither the bare MWCNTs nor bare Mn-complex were catalytically active. The Mn-complex/MWCNT produced CO at a constant rate for 48 h with a current density of greater than 2.0 mA cm–2 at −0.39 V (vs RHE). The MWCNTs with electron accumulation properties, together with surface adsorbed K+ ions, provided an environment to stabilize CO2 adjacent to the Mn-complex and significantly lowered the overpotential for CO2 reduction in an aqueous solution, and these results were consistent with density functional theory (DFT) calculations. Experiments clarified that the synergetic effect of the MWCNTs and K+ ions was also applicable to Co...

SPIGAN: Privileged Adversarial Learning from Simulation

Abstract: Deep Learning for Computer Vision depends mainly on the source of supervision.Photo-realistic simulators can generate large-scale automatically labeled syntheticdata, but introduce a domain gap negatively impacting performance. We propose anew unsupervised domain adaptation algorithm, called SPIGAN, relying on Sim-ulator Privileged Information (PI) and Generative Adversarial Networks (GAN).We use internal data from the simulator as PI during the training of a target tasknetwork. We experimentally evaluate our approach on semantic segmentation. Wetrain the networks on real-world Cityscapes and Vistas datasets, using only unla-beled real-world images and synthetic labeled data with z-buffer (depth) PI fromthe SYNTHIA dataset. Our method improves over no adaptation and state-of-the-art unsupervised domain adaptation techniques.