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.

A lowering in the cost of batteries and solar PV systems has led to a high uptake of solar battery home systems. In this work, we use the deep deterministic policy gradient algorithm to optimise the charging and discharging behaviour of a battery within such a system. Our approach outputs a continuous action space when it charges and discharges the battery, and can function well in a stochastic environment. We show good performance of this algorithm by lowering the expenditure of a single household on electricity to almost \$1AUD for large batteries across selected weeks within a year.

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.

When a sufficiently large far-field training data is presented, jointly optimizing a multichannel frontend and an end-to-end (E2E) Automatic Speech Recognition (ASR) backend shows promising results. Recent literature has shown traditional beamformer designs, such as MVDR (Minimum Variance Distortionless Response) or fixed beamformers can be successfully integrated as the frontend into an E2E ASR system with learnable parameters. In this work, we propose the self-attention channel combinator (SACC) ASR frontend, which leverages the self-attention mechanism to combine multichannel audio signals in the magnitude spectral domain. Experiments conducted on a multichannel playback test data shows that the SACC achieved a 9.3% WERR compared to a state-of-the-art fixed beamformer-based frontend, both jointly optimized with a ContextNet-based ASR backend. We also demonstrate the connection between the SACC and the traditional beamformers, and analyze the intermediate outputs of the SACC.

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.

Driving is an intuitive task that requires skills, constant alertness and vigilance for unexpected events. The driving task also requires long concentration spans focusing on the entire task for prolonged periods, and sophisticated negotiation skills with other road users, including wild animals. These requirements are particularly important when approaching intersections, overtaking, giving way, merging, turning and while adhering to the vast body of road rules. Modern motor vehicles now include an array of smart assistive and autonomous driving systems capable of subsuming some, most, or in limited cases, all of the driving task. The UK Department of Transport's response to the Safe Use of Automated Lane Keeping System consultation proposes that these systems are tested for compliance with relevant traffic rules. Building these smart automotive systems requires software developers with highly technical software engineering skills, and now a lawyer's in-depth knowledge of traffic legislation as well. These skills are required to ensure the systems are able to safely perform their tasks while being observant of the law. This paper presents an approach for deconstructing the complicated legalese of traffic law and representing its requirements and flow. The approach (de)constructs road rules in legal terminology and specifies them in structured English logic that is expressed as Boolean logic for automation and Lawmaps for visualisation. We demonstrate an example using these tools leading to the construction and validation of a Bayesian Network model. We strongly believe these tools to be approachable by programmers and the general public, and capable of use in developing Artificial Intelligence to underpin motor vehicle smart systems, and in validation to ensure these systems are considerate of the law when making decisions.

Compositional data are non-negative data collected in a rectangular matrix with a constant row sum. Due to the non-negativity the focus is on conditional proportions that add up to 1 for each row. A row of conditional proportions is called an observed budget. Latent budget analysis (LBA) assumes a mixture of latent budgets that explains the observed budgets. LBA is usually fitted to a contingency table, where the rows are levels of one or more explanatory variables and the columns the levels of a response variable. In prospective studies, there is only knowledge about the explanatory variables of individuals and interest goes out to predicting the response variable. Thus, a form of LBA is needed that has the functionality of prediction. Previous studies proposed a constrained neural network (NN) extension of LBA that was hampered by an unsatisfying prediction ability. Here we propose LBA-NN, a feed forward NN model that yields a similar interpretation to LBA but equips LBA with a better ability of prediction. A stable and plausible interpretation of LBA-NN is obtained through the use of importance plots and table, that show the relative importance of all explanatory variables on the response variable. An LBA-NN-K- means approach that applies K-means clustering on the importance table is used to produce K clusters that are comparable to K latent budgets in LBA. Here we provide different experiments where LBA-NN is implemented and compared with LBA. In our analysis, LBA-NN outperforms LBA in prediction in terms of accuracy, specificity, recall and mean square error. We provide open-source software at GitHub.

Statistical learning additions to physically derived mathematical models are gaining traction in the literature. A recent approach has been to augment the underlying physics of the governing equations with data driven Bayesian statistical methodology. Coined statFEM, the method acknowledges a priori model misspecification, by embedding stochastic forcing within the governing equations. Upon receipt of additional data, the posterior distribution of the discretised finite element solution is updated using classical Bayesian filtering techniques. The resultant posterior jointly quantifies uncertainty associated with the ubiquitous problem of model misspecification and the data intended to represent the true process of interest. Despite this appeal, computational scalability is a challenge to statFEM's application to high-dimensional problems typically experienced in physical and industrial contexts. This article overcomes this hurdle by embedding a low-rank approximation of the underlying dense covariance matrix, obtained from the leading order modes of the full-rank alternative. Demonstrated on a series of reaction-diffusion problems of increasing dimension, using experimental and simulated data, the method reconstructs the sparsely observed data-generating processes with minimal loss of information, in both posterior mean and the variance, paving the way for further integration of physical and probabilistic approaches to complex systems.

This is the first part of a 2-part series on the growing importance of teaching Data and AI literacy to our students. This will be included in a module I am teaching at Menlo College but wanted to share the blog to help validate the content before presenting to my students. Apple plans to introduce new iPhone software that uses artificial intelligence (AI) to churn through the vast collection of photos that people have taken with their iPhones to detect and report child sexual abuse. See the Wall Street article "Apple Plans to Have iPhones Detect Child Pornography, Fueling Priva..." for more details on Apple's plan. Apple has a strong history of working to protect its customers' privacy.

Dating app Tinder has launched a new tool called Explore that lets users search for potential matches based on their interests. In Explore, users can discover dates who share a love for'every mood and activity', such as gaming, music, food and – for those who want to form a'power couple' – entrepreneurship. Explore, found as a separate tab within the Tinder app, expands on previous filters that had helped users find a date – age, location and sexuality. By giving users the option to navigate through profiles arranged by interest, Tinder is giving users more control over who they meet, according to the company. Explore is now rolling out for users in the UK, the US, Australia and New Zealand, and will be available globally by mid-October.