Private companies are testing new power systems for longer rover missions and future human lunar habitats. Breakthroughs, discoveries, and DIY tips sent every weekday. A future lunar lander bound for the dark side of the moon will carry along a piece of equipment that could make these missions a little bit brighter. The lander in question is operated by Firefly Aerospace, the first commercial company to successfully land and operate spacecraft on the moon. A LightPort wireless power receiver will be mounted atop the Firefly Blue Ghost lander's upper deck.Developed by Canadian aerospace startup Volta Space Technologies, the cargo plays a key role in Volta's ultimate goal: establishing a network of satellites that can wirelessly beam solar power to spacecraft on the lunar surface.

Vision-language pre-training (VLP) has recently proven highly effective for various uni- and multi-modal downstream applications. However, most existing end-to-end VLP methods use high-resolution image-text box data to perform well on fine-grained region-level tasks, such as object detection, segmentation, and referring expression comprehension. Unfortunately, such high-resolution images with accurate bounding box annotations are expensive to collect and use for supervision at scale. In this work, we propose VoLTA (Vision-Language Transformer with weakly-supervised local-feature Alignment), a new VLP paradigm that only utilizes image-caption data but achieves fine-grained region-level image understanding, eliminating the use of expensive box annotations. VoLTA adopts graph optimal transport-based weakly-supervised alignment on local image patches and text tokens to germinate an explicit, self-normalized, and interpretable low-level matching criterion. In addition, VoLTA pushes multi-modal fusion deep into the uni-modal backbones during pre-training and removes fusion-specific transformer layers, further reducing memory requirements. Extensive experiments on a wide range of vision- and vision-language downstream tasks demonstrate the effectiveness of VoLTA on fine-grained applications without compromising the coarse-grained downstream performance, often outperforming methods using significantly more caption and box annotations.

Previous question-answer pair generation methods aimed to produce fluent and meaningful question-answer pairs but tend to have poor diversity. Recent attempts addressing this issue suffer from either low model capacity or overcomplicated architecture. Furthermore, they overlooked the problem where the controllability of their models is highly dependent on the input. In this paper, we propose a model named VOLTA that enhances generative diversity by leveraging the Variational Autoencoder framework with a shared backbone network as its encoder and decoder. In addition, we propose adding InfoGAN-style latent codes to enable input-independent controllability over the generation process. We perform comprehensive experiments and the results show that our approach can significantly improve diversity and controllability over state-of-the-art models.

In clinical practice, many diagnosis tasks rely on the identification of cells in histopathology images. While supervised machine learning techniques require labels, providing manual cell annotations is time-consuming due to the large number of cells. In this paper, we propose a self-supervised framework (VOLTA) for cell representation learning in histopathology images using a novel technique that accounts for the cell's mutual relationship with its environment for improved cell representations. We subjected our model to extensive experiments on the data collected from multiple institutions around the world comprising of over 700,000 cells, four cancer types, and cell types ranging from three to six categories for each dataset. The results show that our model outperforms the state-of-the-art models in cell representation learning. To showcase the potential power of our proposed framework, we applied VOLTA to ovarian and endometrial cancers with very small sample sizes (10-20 samples) and demonstrated that our cell representations can be utilized to identify the known histotypes of ovarian cancer and provide novel insights that link histopathology and molecular subtypes of endometrial cancer. Unlike supervised deep learning models that require large sample sizes for training, we provide a framework that can empower new discoveries without any annotation data in situations where sample sizes are limited.

Volta Inc. an industry-leading electric vehicle ("EV") charging network powering vehicles and commerce, announced its partnership with the City of Hoboken to install 25 conveniently located public EV charging stalls for the City's residents, annual visitors, and commuters over the next 18 months. The addition of the new Volta charging stalls will more than double the number of public EV charging ports available in Hoboken, and Volta and the City of Hoboken may partner to bring even more chargers to the area in the future. The collaboration is a model for how EV charging infrastructure can be efficiently deployed within densely populated urban areas to maximize economic, health, and climate benefits. Volta goes beyond charging as the only EV charging company that directly integrates an eye-catching digital media network into its public charging stations, capturing consumers' attention as they shop and dine at local businesses and walk to entertainment venues, work, and home. Adding 25 new Volta charging stalls and 50 digital media screens in Hoboken expands Volta Media Network's impressions by nearly 20 percent within the New York, NY designated market area (DMA), unlocking additional reach within the most highly ranked DMA for Volta's advertising customers.

Volta Inc. ("Volta"), the industry leader in commerce-centric electric vehicle ("EV") charging, announced its expansion into the European market, with an initial focus on Germany, Austria, Switzerland, and France. The announcement was made at the NOAH Conference in Zurich. With unique charging stations that feature high-impact, large-format digital screens located near the entrances of premier commercial locations, Volta's network is among the most utilized in the U.S. For consumers, Volta provides seamless, reliable charging that complements their daily lives and routines. For site partners, the eye-catching displays and premium station locations help drive business by attracting more customers for longer periods of time. For advertisers, Volta stations double as an innovative, digital out-of-home advertising platform, allowing brands to reach shoppers seconds before they enter a store to make a purchase.

A smart cushion that keeps tabs on your furry friend's fitness was all the rage last year, as were robot petsitters, telemedical apps for cats and dogs, and other tech-centric pet accessories. But leave it to Volta, a Milan designer of connected home appliances and industrial apps, to reinvigorate the category with a pet bowl -- Mookkie -- it says is driven by artificial intelligence (AI). Mookkie, which will make its formal debut at the 2019 Consumer Electronics Show in Las Vegas next week, features a built-in camera with a fisheye lens, an IR sensor, and an embedded circuit board that automatically detects pets. If a recognizable face -- or muzzle, rather -- comes within range of its sensors, a mechanical panel rotates open to reveal the feed compartment. Better still, Volta says that Mookkie's embedded deep neural network can distinguish between individual pets -- preventing, say, Fido from overfeeding, or a wild animal from stealing your good boy's dinner.

NVIDIA's unveiled a new GPU for AI advancement that has claimed to be the fastest it's build. The TITAN V is powered with its new Volta architecture and goals end-user PCs. NVIDIA stated the launch can help researchers explore AI and create technology. All this equates into system learning functionality that NVIDIA asserts could be your greatest readily available for work station PCs. The card's oriented towards programmers who often utilize computing systems and AI. TITAN promises to offer a more efficient and speedier workflow, enabling developers without needing for their hardware to catch up to iterate.

In NVIDIA's Q1 2019 quarter, the company once again exceeded expectations, reporting a 66% growth in total revenue, including 71% growth in its red-hot datacenter business (reaching $701M for the quarter). For NVIDIA, the "Datacenter" segment includes High-Performance Computing (HPC), datacenter-hosted graphics, and AI acceleration. While that is certainly an impressive growth rate, it is smaller than the 2-3x year-over-year growth the company has enjoyed over the last few years. This raises a few interesting questions we will examine here. Is this slow-down in growth a sea change or just the law of large numbers catching up with the business?

There is no question right now that if you have a big computing job in either high performance computing – the colloquial name for traditional massively parallel simulation and modeling applications – or in machine learning – the set of statistical analysis routines with feedback loops that can do identification and transformation tasks that used to be solely the realm of humans – then an Nvidia GPU accelerator is the engine of choice to run that work at the best efficiency. It is usually difficult to make such clean proclamations in the IT industry, with so many different kinds of compute available. But Nvidia is in a unique position, and one that it has earned through more than a decade of intense engineering, where it really does not have effective competition in the compute areas where it plays. Parallel routines written in C, C, or Fortran were offloaded from CPUs to GPUs in the first place because the CPUs did not have sufficient memory bandwidth to handle these routines. It was convenient, perhaps, that the parallel engine used to produce graphics for games and virtualization could be tweaked so it could do raw compute for simulation and modeling.