Machine vision quality assurance systems have excelled at automating the location, identification, and inspection of manufactured components through computational image analysis. But when the component is part of a larger assembly, a complex package, or a kit--such as an automotive assembly or surgical intubation kit--defects, random product placement, variations in lighting, and other factors can quickly overwhelm a traditional machine vision system. For this reason, final inspection of assemblies, packages, and kits is usually conducted manually, to the detriment of overall quality and productivity. While manual operators typically excel at inspecting complex assemblies, by comparing multiple attached or connected components to automated quality inspection solutions, it's hard for operators to stay sharp. Studies show most operators can only focus on a single task for 15 to 20 minutes at a time.

Current vision algorithms are largely designed for use in clear conditions, deep learning using convoluted neural networks is now being harnessed to improve visual performance in adverse weather. Capturing high-quality photos is easier than ever, as filters and image-adjustment tools can enhance images. Yet cameras still struggle to provide a clear image in bad weather, especially in extreme conditions such as heavy rain, fog, or poor lighting at night. Objects in a scene can become hard to see or even invisible, especially when they are far from the lens, and colors are often dulled. "In rain and snow, you also have motion blur because they are moving," says Dengxin Dai, a computer vision lecturer at ETH Zurich in Switzerland who coordinated a workshop on all-weather vision at the Conference for Computer Vision and Pattern Recognition (CVPR 2019) in Long Beach, CA. "So the geometry of an object might also get distorted."

Until 2019, it has been the case that if you come across several paragraphs of text on a consistent topic with consistent subjects, you can assume that text was written or structured by a human being. That is no longer true. Over the past year, AI researchers designed computer programs with the ability to generate multi-paragraph stories that remain fairly coherent throughout. As we explain in the video above, these programs create passages of text that seem like they were written by someone who is fluent in the language but possibly faking their knowledge. I'm not sure we needed an automated version of that person, but here it is: Depending on how you look at it, this technology is a powerful bullshit machine or a promising tool for artists. So far, the creative uses seem to outnumber the malicious ones, but it's not difficult to imagine how text-fakes could cause harm, especially since these models have been widely shared and are now deployable by anyone with basic know-how.

Cryo-EM Single Particle Analysis workflows require from tens of thousands of high-quality particle projections to unveil the three-dimensional structure of macromolecules. Conventional methods for automatic particle picking tend to suffer from high false-positive rates, hurdling the reconstruction process. One common cause of this problem is the presence of carbon and different types of high-contrast contaminations. In order to overcome this limitation, we have developed MicrographCleaner, a deep learning package designed to discriminate which regions of micrographs are suitable for particle picking and which are not in an automatic fashion. MicrographCleaner implements a U-net-like deep learning model trained on a manually curated dataset compiled from over five hundred micrographs.

Artificial Intelligence has progressed immensely in the past few years. From being just a fiction context to penetrating into the regular lives of people, AI has brought transformation in several ways. Such advancements are an output of various factors that include the application of new statistical approaches and enhanced computing powers. However, according to 2017 report by DeepMind, a Perspective in the journal Neuron, argues that people often discount the contribution and use of ideas from experimental and theoretical neuroscience. The DeepMind report's researchers believe that drawing inspiration from neuroscience in AI research is important for two reasons.

Beidi Chen and Tharun Medini, graduate students in computer science at Rice University, helped develop SLIDE, an algorithm for training deep neural networks without graphics processing units. Rice University computer scientists have overcome a major obstacle in the burgeoning artificial intelligence industry by showing it is possible to speed up deep learning technology without specialized acceleration hardware like GPUs. Computer scientists from Rice, supported by collaborators from Intel, will present their results today at the Austin Convention Center as a part of the machine learning systems conference MLSys. Many companies are investing heavily in GPUs and other specialized hardware to implement deep learning, a powerful form of artificial intelligence that's behind digital assistants like Alexa and Siri, facial recognition, product recommendation systems and other technologies. For example, Nvidia, the maker of the industry's gold-standard Tesla V100 Tensor Core GPUs, recently reported a 41% increase in its fourth quarter revenues compared with the previous year.

This article is part of our reviews of AI research papers, a series of posts that explore the latest findings in artificial intelligence. Deep learning, the main innovation that has renewed interest in artificial intelligence in the past years, has helped solve many critical problems in computer vision, natural language processing, and speech recognition. However, as the deep learning matures and moves from hype peak to its trough of disillusionment, it is becoming clear that it is missing some fundamental components. This is a reality that many of the pioneers of deep learning and its main component, artificial neural networks, have acknowledged in various AI conferences in the past year. Geoffrey Hinton, Yann LeCun, and Yoshua Bengio, the three "godfathers of deep learning," have all spoken about the limits of neural networks.

Efficient and effective learning is one of the ultimate goals of the deep reinforcement learning (DRL), although the compromise has been made in most of the time, especially for the application of robot manipulations. Learning is always expensive for robot manipulation tasks and the learning effectiveness could be affected by the system uncertainty. In order to solve above challenges, in this study, we proposed a simple but powerful reward shaping method, namely Dense2Sparse. It combines the advantage of fast convergence of dense reward and the noise isolation of the sparse reward, to achieve a balance between learning efficiency and effectiveness, which makes it suitable for robot manipulation tasks. We evaluated our Dense2Sparse method with a series of ablation experiments using the state representation model with system uncertainty. The experiment results show that the Dense2Sparse method obtained higher expected reward compared with the ones using standalone dense reward or sparse reward, and it also has a superior tolerance of system uncertainty.

Importance: Following a century of increase, life expectancy in the United States has stagnated and begun to decline in recent decades. Using satellite images and street view images prior work has demonstrated associations of the built environment with income, education, access to care and health factors such as obesity. However, assessment of learned image feature relationships with variation in crude mortality rate across the United States has been lacking. Objective: Investigate prediction of county-level mortality rates in the U.S. using satellite images. Design: Satellite images were extracted with the Google Static Maps application programming interface for 430 counties representing approximately 68.9% of the US population. A convolutional neural network was trained using crude mortality rates for each county in 2015 to predict mortality. Learned image features were interpreted using Shapley Additive Feature Explanations, clustered, and compared to mortality and its associated covariate predictors. Main Outcomes and Measures: County mortality was predicted using satellite images. Results: Predicted mortality from satellite images in a held-out test set of counties was strongly correlated to the true crude mortality rate (Pearson r=0.72). Learned image features were clustered, and we identified 10 clusters that were associated with education, income, geographical region, race and age. Conclusion and Relevance: The application of deep learning techniques to remotely-sensed features of the built environment can serve as a useful predictor of mortality in the United States. Tools that are able to identify image features associated with health-related outcomes can inform targeted public health interventions.

Recent research has proposed the lottery ticket hypothesis, suggesting that for a deep neural network, there exist trainable sub-networks performing equally or better than the original model with commensurate training steps. While this discovery is insightful, finding proper sub-networks requires iterative training and pruning. The high cost incurred limits the applications of the lottery ticket hypothesis. We show there exists a subset of the aforementioned sub-networks that converge significantly faster during the training process and thus can mitigate the cost issue. We conduct extensive experiments to show such sub-networks consistently exist across various model structures for a restrictive setting of hyperparameters ($e.g.$, carefully selected learning rate, pruning ratio, and model capacity). As a practical application of our findings, we demonstrate that such sub-networks can help in cutting down the total time of adversarial training, a standard approach to improve robustness, by up to 49\% on CIFAR-10 to achieve the state-of-the-art robustness.