Pest and disease control plays a key role in agriculture since the damage caused by these agents are responsible for a huge economic loss every year. Based on this assumption, we create an algorithm capable of detecting rust (Hemileia vastatrix) and leaf miner (Leucoptera coffeella) in coffee leaves (Coffea arabica) and quantify disease severity using a mobile application as a high-level interface for the model inferences. We used different convolutional neural network architectures to create the object detector, besides the OpenCV library, k-means, and three treatments: the RGB and value to quantification, and the AFSoft software, in addition to the analysis of variance, where we compare the three methods. The results show an average precision of 81,5% in the detection and that there was no significant statistical difference between treatments to quantify the severity of coffee leaves, proposing a computationally less costly method. The application, together with the trained model, can detect the pest and disease over different image conditions and infection stages and also estimate the disease infection stage.

This paper presents a supervised learning method to generate continuous cost-to-go functions of non-holonomic systems directly from the workspace description. Supervision from informative examples reduces training time and improves network performance. The manifold representing the optimal trajectories of a non-holonomic system has high-curvature regions which can not be efficiently captured with uniform sampling. To address this challenge, we present an adaptive sampling method which makes use of sampling-based planners along with local, closed-form solutions to generate training samples. The cost-to-go function over a specific workspace is represented as a neural network whose weights are generated by a second, higher order network. The networks are trained in an end-to-end fashion. In our previous work, this architecture was shown to successfully learn to generate the cost-to-go functions of holonomic systems using uniform sampling. In this work, we show that uniform sampling fails for non-holonomic systems. However, with the proposed adaptive sampling methodology, our network can generate near-optimal trajectories for non-holonomic systems while avoiding obstacles. Experiments show that our method is two orders of magnitude faster compared to traditional approaches in cluttered environments.

We present graph-based translation models which translate source graphs into target strings. Source graphs are constructed from dependency trees with extra links so that non-syntactic phrases are connected. Inspired by phrase-based models, we first introduce a translation model which segments a graph into a sequence of disjoint subgraphs and generates a translation by combining subgraph translations left-to-right using beam search. However, similar to phrase-based models, this model is weak at phrase reordering. Therefore, we further introduce a model based on a synchronous node replacement grammar which learns recursive translation rules. We provide two implementations of the model with different restrictions so that source graphs can be parsed efficiently. Experiments on Chinese--English and German--English show that our graph-based models are significantly better than corresponding sequence- and tree-based baselines.

Most American retail banks are designed the same way: Customers must pass several desks set aside for loan and mortgage officers before they can talk to a customer representative. I only step inside a bank a few times each year, but even pre-pandemic, I can't remember the last time I saw someone sitting at one of those desks. Everyone I know who's obtained a home or business loan in the recent past started with an online application process. For this morning's column, Alex Wilhelm interviewed Dave Girouard, CEO of Upstart, an AI-powered fintech lender that expects to see growth increase 114% this year. A forecast like that suggests that retail banks have gotten comfortable with using automated tools to calculate risk, which may help explain all the empty desks at my local branch.

The illegal animal trade is a global scourge but a lucrative one, worth $8 to 10 billion annually, according to the United Nations Office on Drugs and Crime (UNODC) -- trailing only human, drug and weapons trafficking in value. With so much money to be made, conservationists and wildlife rangers face overwhelming odds against well-organized poaching operations fueled by incessant demand for illicit animal products. The results of this protracted conflict have been nothing short of devastating for the species caught in the middle. At the start of the 20th century, more than 100,000 tigers are estimated to have roamed throughout Southeast Asia. Today, due to a combination of habitat loss and aggressive poaching, fewer than 4,000 currently remain in the wild.

Britain's military will unveil a shift towards more lethal, hi-tech and drone-enabled warfare on Monday as ministers and chiefs attempt to stave off criticism of impending cuts in the size of the armed forces. The plan will be highlighted in a defence command paper setting out the military's ambitions for the next five years and confirming a cut in the size of the army to an anticipated 72,500 troops, and a string of other savings as day-to-day defence budgets are squeezed. Ben Wallace, the defence secretary, said on Friday it was time to end "the Top Trumps game of numbers" because previous reviews that had emphasised size had left the military with "lots of ships that are tied up and not available, or lots of regiments". Instead, ministers and service chiefs will highlight how forces such as the Royal Marines could use a mobile phone app to locate friends and enemies on a battlefield while using Ghost drones, 6ft-long single-blade helicopter-like devices that can highlight and even fire at targets. Gen Sir Nick Carter, the head of the armed forces, said that "rather than focus on size and shape, I would focus on lethality, the relevance, the resilience and the readiness of our army and our armed forces."

We develop an algorithm to train individually fair learning-to-rank (LTR) models. The proposed approach ensures items from minority groups appear alongside similar items from majority groups. This notion of fair ranking is based on the definition of individual fairness from supervised learning and is more nuanced than prior fair LTR approaches that simply ensure the ranking model provides underrepresented items with a basic level of exposure. The crux of our method is an optimal transport-based regularizer that enforces individual fairness and an efficient algorithm for optimizing the regularizer. We show that our approach leads to certifiably individually fair LTR models and demonstrate the efficacy of our method on ranking tasks subject to demographic biases. Information retrieval (IR) systems are everywhere in today's digital world, and ranking models are integral parts of many IR systems. In light of their ubiquity, issues of algorithmic bias and unfairness in ranking models have come to the fore of the public's attention. In many applications, the items to be ranked are individuals, so algorithmic biases in the output of ranking models directly affect people's lives. For example, gender bias in job search engines directly affect the career success of job applicants (Dastin, 2018).

In this paper we provide results on using integer programming (IP) and constraint programming (CP) to search for sets of mutually orthogonal latin squares (MOLS). Both programming paradigms have previously successfully been used to search for MOLS, but solvers for IP and CP solvers have significantly improved in recent years and data on how modern IP and CP solvers perform on the MOLS problem is lacking. Using state-of-the-art solvers as black boxes we were able to quickly find pairs of MOLS (or prove their nonexistence) in all orders up to ten. Moreover, we improve the effectiveness of the solvers by formulating an extended symmetry breaking method as well as an improvement to the straightforward CP encoding. We also analyze the effectiveness of using CP and IP solvers to search for triples of MOLS, compare our timings to those which have been previously published, and estimate the running time of using this approach to resolve the longstanding open problem of determining the existence of a triple of MOLS of order ten.

Skeletal motion plays a vital role in human activity recognition as either an independent data source or a complement. The robustness of skeleton-based activity recognizers has been questioned recently, which shows that they are vulnerable to adversarial attacks when the full-knowledge of the recognizer is accessible to the attacker. However, this white-box requirement is overly restrictive in most scenarios and the attack is not truly threatening. In this paper, we show that such threats do exist under black-box settings too. To this end, we propose the first black-box adversarial attack method BASAR. Through BASAR, we show that adversarial attack is not only truly a threat but also can be extremely deceitful, because on-manifold adversarial samples are rather common in skeletal motions, in contrast to the common belief that adversarial samples only exist off-manifold. Through exhaustive evaluation and comparison, we show that BASAR can deliver successful attacks across models, data, and attack modes. Through harsh perceptual studies, we show that it achieves effective yet imperceptible attacks. By analyzing the attack on different activity recognizers, BASAR helps identify the potential causes of their vulnerability and provides insights on what classifiers are likely to be more robust against attack.

The next generation of wireless networks, dubbed 5G, will have more capacity, faster speeds, and lower latency than its predecessor 4G. As a result, it's expected to bring technologies like augmented reality, self-driving cars, data-crunching Internet of Things devices, and even smart cities closer to the mainstream than ever before. Deeply entwined with cloud computing, 5G is expected to be the backbone of so many future products and services that it has the potential to power economic growth for decades to come, analysts predict. At the moment, 5G networks are still being rolled out by wireless carriers, and the public has yet to fully realize its benefits. But there are plenty of opportunities for startups and major companies alike, including in partnering with wireless carriers, deploying private and enterprise 5G networks, and developing 5G-enabled applications.