Developers of text generation models rely on automated evaluation metrics as a stand-in for slow and expensive manual evaluations. However, image captioning metrics have struggled to give accurate learned estimates of the semantic and pragmatic success of output text. We address this weakness by introducing the first discourse-aware learned generation metric for evaluating image descriptions. Our approach is inspired by computational theories of discourse for capturing information goals using coherence. We present a dataset of image$\unicode{x2013}$description pairs annotated with coherence relations. We then train a coherence-aware metric on a subset of the Conceptual Captions dataset and measure its effectiveness$\unicode{x2014}$its ability to predict human ratings of output captions$\unicode{x2014}$on a test set composed of out-of-domain images. We demonstrate a higher Kendall Correlation Coefficient for our proposed metric with the human judgments for the results of a number of state-of-the-art coherence-aware caption generation models when compared to several other metrics including recently proposed learned metrics such as BLEURT and BERTScore.

Berlin, Germany: A global early warning centre in Berlin will use artificial intelligence to predict the next pandemic. But AI's already helped us fight COVID-19. If artificial intelligence is the future, then the future is now. This pandemic has shown us just how fast artificial intelligence (AI) works and what it can do in so many different ways. It may also help us predict the next pandemic.

Association Rules is one of the very important concepts of machine learning being used in market basket analysis. This course covers the working Principle of Association Mining and its various concepts like Support, Confidence, and Life in a very simplified manner. All of these algorithms has been explained by taking working examples. Parteek Bhatia is Professor in the Department of Computer Science and Engineering and Former Associate Dean of Student Affairs at Thapar Institute of Engineering and Technology, Patiala. At present he is on sabbatical at Tel Aviv University, Israel and acting as Visiting Professor at LAMBDA Lab, TAU.

Striking a blow to patent applicants seeking to assert inventorship by artificial intelligence ("AI") systems, the U.S. District Court for the Eastern District of Virginia ruled on September 3, 2021 that an AI machine cannot qualify as an "inventor" under the Patent Act. The fight is now expected to move to the Federal Circuit on appeal. Proskauer has been closely monitoring the quickly-developing legal treatment of AI systems, especially in view of their implications for life sciences patents. AI's presence in life sciences innovation is well established, for example, to predict biological targets of prospective drug molecules and to identify drug design candidates (among many other applications). As we reported in August, two countries--Australia and South Africa--have already permitted AI systems to qualify as "inventors" in patent applications. However, hope for a worldwide trend have been dashed, at least for now.

Intelligent AI provides a 360-degree view of risk, with more than 300 datasets, including AI, IoT, Satellite, NatCat, and Open Data. "Intelligent AI offers an exceptionally relevant use case for the industry right now," said Chris Newman, global managing director at ACORD. "Their presentation demonstrated the art of the possible in what can be done with data in the insurance commercial lines space." Apart from the cash prize, the firm will be featured in an ACORD-promoted webinar to present its innovation. "I'd like to thank ACORD and the AIIC judges for this great opportunity," said Anthony Peake, chief executive officer at Intelligent AI. "We have had a fantastic first year at Intelligent AI and are excited for the future. We think real-time data and digital twins are going to be an industry game-changer for both insurers and commercial customers."

Both image-caption pairs and translation pairs provide the means to learn deep representations of and connections between languages. We use both types of pairs in MURAL (MUltimodal, MUltitask Representations Across Languages), a dual encoder that solves two tasks: 1) image-text matching and 2) translation pair matching. By incorporating billions of translation pairs, MURAL extends ALIGN (Jia et al. PMLR'21)--a state-of-the-art dual encoder learned from 1.8 billion noisy image-text pairs. When using the same encoders, MURAL's performance matches or exceeds ALIGN's cross-modal retrieval performance on well-resourced languages across several datasets. More importantly, it considerably improves performance on under-resourced languages, showing that text-text learning can overcome a paucity of image-caption examples for these languages. On the Wikipedia Image-Text dataset, for example, MURAL-base improves zero-shot mean recall by 8.1% on average for eight under-resourced languages and by 6.8% on average when fine-tuning. We additionally show that MURAL's text representations cluster not only with respect to genealogical connections but also based on areal linguistics, such as the Balkan Sprachbund.

State-of-the-art contextual embeddings are obtained from large language models available only for a few languages. For others, we need to learn representations using a multilingual model. There is an ongoing debate on whether multilingual embeddings can be aligned in a space shared across many languages. The novel Orthogonal Structural Probe (Limisiewicz and Mare\v{c}ek, 2021) allows us to answer this question for specific linguistic features and learn a projection based only on mono-lingual annotated datasets. We evaluate syntactic (UD) and lexical (WordNet) structural information encoded inmBERT's contextual representations for nine diverse languages. We observe that for languages closely related to English, no transformation is needed. The evaluated information is encoded in a shared cross-lingual embedding space. For other languages, it is beneficial to apply orthogonal transformation learned separately for each language. We successfully apply our findings to zero-shot and few-shot cross-lingual parsing.

We present ReasonBert, a pre-training method that augments language models with the ability to reason over long-range relations and multiple, possibly hybrid contexts. Unlike existing pre-training methods that only harvest learning signals from local contexts of naturally occurring texts, we propose a generalized notion of distant supervision to automatically connect multiple pieces of text and tables to create pre-training examples that require long-range reasoning. Different types of reasoning are simulated, including intersecting multiple pieces of evidence, bridging from one piece of evidence to another, and detecting unanswerable cases. We conduct a comprehensive evaluation on a variety of extractive question answering datasets ranging from single-hop to multi-hop and from text-only to table-only to hybrid that require various reasoning capabilities and show that ReasonBert achieves remarkable improvement over an array of strong baselines. Few-shot experiments further demonstrate that our pre-training method substantially improves sample efficiency.

Motivated by suggested question generation in conversational news recommendation systems, we propose a model for generating question-answer pairs (QA pairs) with self-contained, summary-centric questions and length-constrained, article-summarizing answers. We begin by collecting a new dataset of news articles with questions as titles and pairing them with summaries of varying length. This dataset is used to learn a QA pair generation model producing summaries as answers that balance brevity with sufficiency jointly with their corresponding questions. We then reinforce the QA pair generation process with a differentiable reward function to mitigate exposure bias, a common problem in natural language generation. Both automatic metrics and human evaluation demonstrate these QA pairs successfully capture the central gists of the articles and achieve high answer accuracy.

In recent years, distributed optimization is proven to be an effective approach to accelerate training of large scale machine learning models such as deep neural networks. With the increasing computation power of GPUs, the bottleneck of training speed in distributed training is gradually shifting from computation to communication. Meanwhile, in the hope of training machine learning models on mobile devices, a new distributed training paradigm called ``federated learning'' has become popular. The communication time in federated learning is especially important due to the low bandwidth of mobile devices. While various approaches to improve the communication efficiency have been proposed for federated learning, most of them are designed with SGD as the prototype training algorithm. While adaptive gradient methods have been proven effective for training neural nets, the study of adaptive gradient methods in federated learning is scarce. In this paper, we propose an adaptive gradient method that can guarantee both the convergence and the communication efficiency for federated learning.