This thesis tackles the subject of spatio-temporal forecasting with deep learning. The motivating application at Electricity de France (EDF) is short-term solar energy forecasting with fisheye images. We explore two main research directions for improving deep forecasting methods by injecting external physical knowledge. The first direction concerns the role of the training loss function. We show that differentiable shape and temporal criteria can be leveraged to improve the performances of existing models. We address both the deterministic context with the proposed DILATE loss function and the probabilistic context with the STRIPE model. Our second direction is to augment incomplete physical models with deep data-driven networks for accurate forecasting. For video prediction, we introduce the PhyDNet model that disentangles physical dynamics from residual information necessary for prediction, such as texture or details. We further propose a learning framework (APHYNITY) that ensures a principled and unique linear decomposition between physical and data-driven components under mild assumptions, leading to better forecasting performances and parameter identification.

Multimodal named entity recognition and relation extraction (MNER and MRE) is a fundamental and crucial branch in information extraction. However, existing approaches for MNER and MRE usually suffer from error sensitivity when irrelevant object images incorporated in texts. To deal with these issues, we propose a novel Hierarchical Visual Prefix fusion NeTwork (HVPNeT) for visual-enhanced entity and relation extraction, aiming to achieve more effective and robust performance. Specifically, we regard visual representation as pluggable visual prefix to guide the textual representation for error insensitive forecasting decision. We further propose a dynamic gated aggregation strategy to achieve hierarchical multi-scaled visual features as visual prefix for fusion. Extensive experiments on three benchmark datasets demonstrate the effectiveness of our method, and achieve state-of-the-art performance. Code is available in https://github.com/zjunlp/HVPNeT.

Nonconvex regularization has been popularly used in low-rank matrix learning. However, extending it for low-rank tensor learning is still computationally expensive. To address this problem, we develop an efficient solver for use with a nonconvex extension of the overlapped nuclear norm regularizer. Based on the proximal average algorithm, the proposed algorithm can avoid expensive tensor folding/unfolding operations. A special "sparse plus low-rank" structure is maintained throughout the iterations, and allows fast computation of the individual proximal steps. Empirical convergence is further improved with the use of adaptive momentum. We provide convergence guarantees to critical points on smooth losses and also on objectives satisfying the Kurdyka-{\L}ojasiewicz condition. While the optimization problem is nonconvex and nonsmooth, we show that its critical points still have good statistical performance on the tensor completion problem. Experiments on various synthetic and real-world data sets show that the proposed algorithm is efficient in both time and space and more accurate than the existing state-of-the-art.

This survey article assesses and compares existing critiques of current fairness-enhancing technical interventions in machine learning (ML) that draw from a range of non-computing disciplines, including philosophy, feminist studies, critical race and ethnic studies, legal studies, anthropology, and science and technology studies. It bridges epistemic divides in order to offer an interdisciplinary understanding of the possibilities and limits of hegemonic computational approaches to ML fairness for producing just outcomes for society's most marginalized. The article is organized according to nine major themes of critique wherein these different fields intersect: 1) how "fairness" in AI fairness research gets defined; 2) how problems for AI systems to address get formulated; 3) the impacts of abstraction on how AI tools function and its propensity to lead to technological solutionism; 4) how racial classification operates within AI fairness research; 5) the use of AI fairness measures to avoid regulation and engage in ethics washing; 6) an absence of participatory design and democratic deliberation in AI fairness considerations; 7) data collection practices that entrench "bias," are non-consensual, and lack transparency; 8) the predatory inclusion of marginalized groups into AI systems; and 9) a lack of engagement with AI's long-term social and ethical outcomes. Drawing from these critiques, the article concludes by imagining future ML fairness research directions that actively disrupt entrenched power dynamics and structural injustices in society.

Researchers have developed a new method that uses artificial intelligence to analyze animal behavior. This opens the door to longer-term in-depth studies in the field of behavioral science--while also helping to improve animal welfare. The method is already being tested at Zurich Zoo. Researchers engaged in animal behavior studies often rely on hours upon hours of video footage which they manually analyze. Usually, this requires researchers to work their way through recordings spanning several weeks or months, laboriously noting down observations on the animals' behavior.

Today, over 2/3 of the people on earth do not have access to radiologists. The are big disparities between counties and within countries. Some countries like the US have tens of thousands of radiologists whereas 14 African countries have no radiologists at all. In India there is approximately one radiologist for every 100,000 people whereas in the US there is one radiologist for every 10,000 people. There are also disparities within countries.

Euphemisms have not received much attention in natural language processing, despite being an important element of polite and figurative language. Euphemisms prove to be a difficult topic, not only because they are subject to language change, but also because humans may not agree on what is a euphemism and what is not. Nevertheless, the first step to tackling the issue is to collect and analyze examples of euphemisms. We present a corpus of potentially euphemistic terms (PETs) along with example texts from the GloWbE corpus. Additionally, we present a subcorpus of texts where these PETs are not being used euphemistically, which may be useful for future applications. We also discuss the results of multiple analyses run on the corpus. Firstly, we find that sentiment analysis on the euphemistic texts supports that PETs generally decrease negative and offensive sentiment. Secondly, we observe cases of disagreement in an annotation task, where humans are asked to label PETs as euphemistic or not in a subset of our corpus text examples. We attribute the disagreement to a variety of potential reasons, including if the PET was a commonly accepted term (CAT).

ClearScore is searching for Data Scientist. This is an amazing opportunity to engage with a rich data set on 15 million users. We have captured every interaction the users have had with our high touch website and native apps, their fully monthly credit files and growing access to insurance, bank account and mobile data as well. Ordinarily a data set this vast would be buried in legacy systems, however ClearScore is only 6 years old and operates a strict'zero legacy' approach that means the data is held in an autoscaling AWS redshift environment. This is an exciting opportunity to work within a fast-paced, rapidly growing team of talented data scientists, helping to build revolutionary experiences that will help millions of users manage and better understand their finances.

In March, a largely overlooked, 90-page Government Accountability Office study revealed something interesting: This summer, the Pentagon is getting a new AI Strategy. Between shaping ethical norms for AI and establishing a new Chief Data and AI Officer, it's clear top brass have big plans for the technology, though the report is light on the details. Released in 2018, the last AI Strategy laid the scaffolding for the U.S. military's high-tech competition with China. But over the past four years one thing has become apparent: The United States needs a balanced approach to AI investment -- one that doesn't simply guard against threats, but also imposes costs on a Chinese force that sees AI as the key to victory. Undoubtedly, a military conflict between the United States and China would be catastrophic, and every effort must be taken to avoid such an outcome through diplomatic means.

Large-scale pre-trained language models have contributed significantly to natural language processing by demonstrating remarkable abilities as few-shot learners. However, their effectiveness depends mainly on scaling the model parameters and prompt design, hindering their implementation in most real-world applications. This study proposes a novel pluggable, extensible, and efficient approach named DifferentiAble pRompT (DART), which can convert small language models into better few-shot learners without any prompt engineering. The main principle behind this approach involves reformulating potential natural language processing tasks into the task of a pre-trained language model and differentially optimizing the prompt template as well as the target label with backpropagation. Furthermore, the proposed approach can be: (i) Plugged to any pre-trained language models; (ii) Extended to widespread classification tasks. A comprehensive evaluation of standard NLP tasks demonstrates that the proposed approach achieves a better few-shot performance. Code is available in https://github.com/zjunlp/DART.