Alongside huge volumes of research on deep learning models in NLP in the recent years, there has been also much work on benchmark datasets needed to track modeling progress. Question answering and reading comprehension have been particularly prolific in this regard, with over 80 new datasets appearing in the past two years. This study is the largest survey of the field to date. We provide an overview of the various formats and domains of the current resources, highlighting the current lacunae for future work. We further discuss the current classifications of ``reasoning types" in question answering and propose a new taxonomy. We also discuss the implications of over-focusing on English, and survey the current monolingual resources for other languages and multilingual resources. The study is aimed at both practitioners looking for pointers to the wealth of existing data, and at researchers working on new resources.

Federated Learning aims to learn machine learning models from multiple decentralized edge devices (e.g. mobiles) or servers without sacrificing local data privacy. Recent Natural Language Processing techniques rely on deep learning and large pre-trained language models. However, both big deep neural and language models are trained with huge amounts of data which often lies on the server side. Since text data is widely originated from end users, in this work, we look into recent NLP models and techniques which use federated learning as the learning framework. Our survey discusses major challenges in federated natural language processing, including the algorithm challenges, system challenges as well as the privacy issues. We also provide a critical review of the existing Federated NLP evaluation methods and tools. Finally, we highlight the current research gaps and future directions.

Agriculture is an essential industry in the both society and economy of a country. However, the pests and diseases cause a great amount of reduction in agricultural production while there is no sufficient guidance for farmers to avoid this disaster. To address this problem, we apply CNNs to plant disease recognition by building a classification model. Within the dataset of 3,642 images of apple leaves [1], We use a pre-trained image classification model Restnet34 based on Convolutional neural network (CNNs) with the Fastai framework in order to save the training time. Overall, the accuracy of classification is 93.765%.

Deep learning has proven effective for various application tasks, but its applicability is limited by the reliance on annotated examples. Self-supervised learning has emerged as a promising direction to alleviate the supervision bottleneck, but existing work focuses on leveraging co-occurrences in unlabeled data for task-agnostic representation learning, as exemplified by masked language model pretraining. In this chapter, we explore task-specific self-supervision, which leverages domain knowledge to automatically annotate noisy training examples for end applications, either by introducing labeling functions for annotating individual instances, or by imposing constraints over interdependent label decisions. We first present deep probabilistic logic(DPL), which offers a unifying framework for task-specific self-supervision by composing probabilistic logic with deep learning. DPL represents unknown labels as latent variables and incorporates diverse self-supervision using probabilistic logic to train a deep neural network end-to-end using variational EM. Next, we present self-supervised self-supervision(S4), which adds to DPL the capability to learn new self-supervision automatically. Starting from an initial seed self-supervision, S4 iteratively uses the deep neural network to propose new self supervision. These are either added directly (a form of structured self-training) or verified by a human expert (as in feature-based active learning). Experiments on real-world applications such as biomedical machine reading and various text classification tasks show that task-specific self-supervision can effectively leverage domain expertise and often match the accuracy of supervised methods with a tiny fraction of human effort.

The ability to edit facial expressions has a wide range of applications in computer graphics. The ideal facial expression editing algorithm needs to satisfy two important criteria. First, it should allow precise and targeted editing of individual facial actions. Second, it should generate high fidelity outputs without artifacts. We build a solution based on StyleGAN, which has been used extensively for semantic manipulation of faces. As we do so, we add to our understanding of how various semantic attributes are encoded in StyleGAN. In particular, we show that a naive strategy to perform editing in the latent space results in undesired coupling between certain action units, even if they are conceptually distinct. For example, although brow lowerer and lip tightener are distinct action units, they appear correlated in the training data. Hence, StyleGAN has difficulty in disentangling them. We allow disentangled editing of such action units by computing detached regions of influence for each action unit, and restrict editing to these regions. We validate the effectiveness of our local editing method through perception experiments conducted with 23 subjects. The results show that our method provides higher control over local editing and produces images with superior fidelity compared to the state-of-the-art methods.

Deep Neural Networks (DNNs) have been shown to be susceptible to memorization or overfitting in the presence of noisily labelled data. For the problem of robust learning under such noisy data, several algorithms have been proposed. A prominent class of algorithms rely on sample selection strategies, motivated by curriculum learning. For example, many algorithms use the `small loss trick' wherein a fraction of samples with loss values below a certain threshold are selected for training. These algorithms are sensitive to such thresholds, and it is difficult to fix or learn these thresholds. Often, these algorithms also require information such as label noise rates which are typically unavailable in practice. In this paper, we propose a data-dependent, adaptive sample selection strategy that relies only on batch statistics of a given mini-batch to provide robustness against label noise. The algorithm does not have any additional hyperparameters for sample selection, does not need any information on noise rates, and does not need access to separate data with clean labels. We empirically demonstrate the effectiveness of our algorithm on benchmark datasets.

Computers have been able to quickly process 2D images for some time. Your cell phone can snap digital photographs and manipulate them in a number of ways. Much more difficult, however, is processing an image in three dimensions, and doing it in a timely manner. The mathematics are more complex, and crunching those numbers, even on a supercomputer, takes time. That's the challenge a group of scientists from the U.S. Department of Energy's (DOE) Argonne National Laboratory is working to overcome.

Retailers around the country rely on a network of nearly 70 independent Coca-Cola bottlers to manufacture and ship cases of liquid refreshments. It's a finely tuned supply chain, and it ultimately serves customers well. But getting a consolidated view into the millions of paper-based billing and shipping documents was a major headache for the network–that is, until an innovative use of computer vision and NLP technology helped digitize it. Starting around 2007, Coca-Cola North America worked to "refranchise" its bottling and shipping operations, which spans 51 production facilities, 350 distribution centers, and involves more than 55,000 employees. The company says the goal of this refranchising effort is to "bring the heart of Coca-Cola back to the local bottler," which in some cases are multi-generational companies more than 100 year's old. A key player in all this is the Coca-Cola Bottler's Sales and Services Company.

Are you a highly motivated researcher with an outstanding track record in Mathematics and its Applications in Science and Engineering? We offer a position at the interface of Dynamics and Deep Learning in the Applied Analysis group of the SACS cluster within the Department of Applied Mathematics (AM) at the University of Twente (UT). The challenge: You will actively develop your mathematical profile and seek connections between fundamental mathematical theory of dynamical systems, nonlinear analysis and the rising area of deep learning for data-driven model discovery. Based on a long-standing expertise and tradition of dynamical systems theory at the UT, well embedded in the Dutch NDNS cluster, our department is looking for a mathematician with a proven expertise in the broad area of dynamical systems, nonlinear analysis or approximation theory for deep neural networks. You show great passion in applying your novel methods to computational neuroscience, inverse problems in imaging or engineering applications driven by physics-informed machine learning for example within the multi-disciplinary research contexts at the UT, like the Digital Society Institute, the Technical Medical Centre or the MESA Institute for Nanotechnology.

This article is part of our reviews of AI research papers, a series of posts that explore the latest findings in artificial intelligence. Even before they speak their first words, human babies develop mental models about objects and people. This is one of the key capabilities that allows us humans to learn to live socially and cooperate (or compete) with each other. But for artificial intelligence, even the most basic behavioral reasoning tasks remain a challenge. Advanced deep learning models can do complicated tasks such as detect people and objects in images, sometimes even better than humans.