Common Deep Metric Learning (DML) datasets specify only one notion of similarity, e.g., two images in the Cars196 dataset are deemed similar if they show the same car model. We argue that depending on the application, users of image retrieval systems have different and changing similarity notions that should be incorporated as easily as possible. Therefore, we present Language-Guided Zero-Shot Deep Metric Learning (LanZ-DML) as a new DML setting in which users control the properties that should be important for image representations without training data by only using natural language. To this end, we propose InDiReCT (Image representations using Dimensionality Reduction on CLIP embedded Texts), a model for LanZ-DML on images that exclusively uses a few text prompts for training. InDiReCT utilizes CLIP as a fixed feature extractor for images and texts and transfers the variation in text prompt embeddings to the image embedding space. Extensive experiments on five datasets and overall thirteen similarity notions show that, despite not seeing any images during training, InDiReCT performs better than strong baselines and approaches the performance of fully-supervised models. An analysis reveals that InDiReCT learns to focus on regions of the image that correlate with the desired similarity notion, which makes it a fast to train and easy to use method to create custom embedding spaces only using natural language.

Fairness in multiwinner elections is studied in varying contexts. For instance, diversity of candidates and representation of voters are both separately termed as being fair. A common denominator to ensure fairness across all such contexts is the use of constraints. However, across these contexts, the candidates selected to satisfy the given constraints may systematically lead to unfair outcomes for historically disadvantaged voter populations as the cost of fairness may be borne unequally. Hence, we develop a model to select candidates that satisfy the constraints fairly across voter populations. To do so, the model maps the constrained multiwinner election problem to a problem of fairly allocating indivisible goods. We propose three variants of the model, namely, global, localized, and inter-sectional. Next, we analyze the model's computational complexity, and we present an empirical analysis of the utility traded-off across various settings of our model across the three variants and discuss the impact of Simpson's paradox using synthetic datasets and a dataset of voting at the United Nations. Finally, we discuss the implications of our work for AI and machine learning, especially for studies that use constraints to guarantee fairness.

Large language models generate complex, open-ended outputs: instead of outputting a class label they write summaries, generate dialogue, or produce working code. In order to asses the reliability of these open-ended generation systems, we aim to identify qualitative categories of erroneous behavior, beyond identifying individual errors. To hypothesize and test for such qualitative errors, we draw inspiration from human cognitive biases -- systematic patterns of deviation from rational judgement. Specifically, we use cognitive biases as motivation to (i) generate hypotheses for problems that models may have, and (ii) develop experiments that elicit these problems. Using code generation as a case study, we find that OpenAI's Codex errs predictably based on how the input prompt is framed, adjusts outputs towards anchors, and is biased towards outputs that mimic frequent training examples. We then use our framework to elicit high-impact errors such as incorrectly deleting files. Our results indicate that experimental methodology from cognitive science can help characterize how machine learning systems behave.

Recent work has demonstrated that pre-trained language models (PLMs) are zero-shot learners. However, most existing zero-shot methods involve heavy human engineering or complicated self-training pipelines, hindering their application to new situations. In this work, we show that zero-shot text classification can be improved simply by clustering texts in the embedding spaces of PLMs. Specifically, we fit the unlabeled texts with a Bayesian Gaussian Mixture Model after initializing cluster positions and shapes using class names. Despite its simplicity, this approach achieves superior or comparable performance on both topic and sentiment classification datasets and outperforms prior works significantly on unbalanced datasets. We further explore the applicability of our clustering approach by evaluating it on 14 datasets with more diverse topics, text lengths, and numbers of classes. Our approach achieves an average of 20% absolute improvement over prompt-based zero-shot learning. Finally, we compare different PLM embedding spaces and find that texts are well-clustered by topics even if the PLM is not explicitly pre-trained to generate meaningful sentence embeddings. This work indicates that PLM embeddings can categorize texts without task-specific fine-tuning, thus providing a new way to analyze and utilize their knowledge and zero-shot learning ability.

The increasing deployment of artificial intelligence (AI) tools to inform decision making across diverse areas including healthcare, employment, social benefits, and government policy, presents a serious risk for disabled people, who have been shown to face bias in AI implementations. While there has been significant work on analysing and mitigating algorithmic bias, the broader mechanisms of how bias emerges in AI applications are not well understood, hampering efforts to address bias where it begins. In this article, we illustrate how bias in AI-assisted decision making can arise from a range of specific design decisions, each of which may seem self-contained and non-biasing when considered separately. These design decisions include basic problem formulation, the data chosen for analysis, the use the AI technology is put to, and operational design elements in addition to the core algorithmic design. We draw on three historical models of disability common to different decision-making settings to demonstrate how differences in the definition of disability can lead to highly distinct decisions on each of these aspects of design, leading in turn to AI technologies with a variety of biases and downstream effects. We further show that the potential harms arising from inappropriate definitions of disability in fundamental design stages are further amplified by a lack of transparency and disabled participation throughout the AI design process. Our analysis provides a framework for critically examining AI technologies in decision-making contexts and guiding the development of a design praxis for disability-related AI analytics. We put forth this article to provide key questions to facilitate disability-led design and participatory development to produce more fair and equitable AI technologies in disability-related contexts.

Transfer of recent advances in deep reinforcement learning to real-world applications is hindered by high data demands and thus low efficiency and scalability. Through independent improvements of components such as replay buffers or more stable learning algorithms, and through massively distributed systems, training time could be reduced from several days to several hours for standard benchmark tasks. However, while rewards in simulated environments are well-defined and easy to compute, reward evaluation becomes the bottleneck in many real-world environments, e.g., in molecular optimization tasks, where computationally demanding simulations or even experiments are required to evaluate states and to quantify rewards. Therefore, training might become prohibitively expensive without an extensive amount of computational resources and time. We propose to alleviate this problem by replacing costly ground-truth rewards with rewards modeled by neural networks, counteracting non-stationarity of state and reward distributions during training with an active learning component. We demonstrate that using our proposed ACRL method (actively learning costly rewards for reinforcement learning), it is possible to train agents in complex real-world environments orders of magnitudes faster. By enabling the application of reinforcement learning methods to new domains, we show that we can find interesting and non-trivial solutions to real-world optimization problems in chemistry, materials science and engineering.

We study an issue commonly seen with graph data analysis: many real-world complex systems involving high-order interactions are best encoded by hypergraphs; however, their datasets often end up being published or studied only in the form of their projections (with dyadic edges). To understand this issue, we first establish a theoretical framework to characterize this issue's implications and worst-case scenarios. The analysis motivates our formulation of the new task, supervised hypergraph reconstruction: reconstructing a real-world hypergraph from its projected graph, with the help of some existing knowledge of the application domain. To reconstruct hypergraph data, we start by analyzing hyperedge distributions in the projection, based on which we create a framework containing two modules: (1) to handle the enormous search space of potential hyperedges, we design a sampling strategy with efficacy guarantees that significantly narrows the space to a smaller set of candidates; (2) to identify hyperedges from the candidates, we further design a hyperedge classifier in two well-working variants that capture structural features in the projection. Extensive experiments validate our claims, approach, and extensions. Remarkably, our approach outperforms all baselines by an order of magnitude in accuracy on hard datasets. Our code and data can be downloaded from bit.ly/SHyRe.

We adopted a Delphi method, which is commonly used when the problem under analysis can benefit from collective and subjective judgments or decisions and when group dynamics do not allow for effective communication (for example, time differences, distance).14 Three of the authors, in weekly meetings, iteratively analyzed the extracted data, resolved ambiguity, and converged onto the final abstraction shown in Table 1. Based on a final data analysis, we made three key observations.

On Saturday, October 29, a Russian fleet on the Black Sea near Sevastopol was attacked by 16 drones--nine in the air and seven in the water. Purportedly launched by Ukraine, no one knows how much damage was done, but video shot by the attacking drones showed that the vessels were unable to avoid being hit. In response to that and other successful attacks, Russia has retaliated with scores of missiles and Iranian-built Shahed-136 drones aimed at electrical and water systems throughout Ukraine. Despite daily reports of lands taken or lands liberated in the nine-month war, the conflict has been largely fought in the air, with artillery shells, rockets, cruise missiles, and, increasingly, drones. Small, cheap, relatively slow-moving, carrying far less of a wallop than a cruise missile or a 500-pound bomb, the Shaheds in particular have bedeviled Ukraine's otherwise excellent air defenses.

Proudly "voted the best place to work" in 2021-2022, Foodics, one of the most promising SaaS companies in MENA, was founded in 2014 in KSA with headquarters in Riyadh and offices in the United Arab Emirates Jordan, Kuwait, Egypt, Pakistan, and the Netherlands. FOODICS is the leading Restaurant-Tech company in MENA and a pioneer in the regional F&B industry. Foodics is undergoing rapid expansion across MENA, Pakistan, Africa and Asia, servicing over 20,000 brands, and has achieved three rounds of funding, with the latest raising $170 million in the largest SaaS funding round in MENA, boosting its innovation capabilities to better serve business owners. We provide a cloud-based point-of-sale SaaS ecosystem with tools that help F&B, and retail businesses start, track and grow. Our customers use Foodics to accept payments, track inventory, monitor sales, process orders, digitize menus, manage employees, create analytics and smart reports, provide secure cloud storage and enable the integration of third-party apps.