This report surveys the landscape of potential security threats from malicious uses of AI, and proposes ways to better forecast, prevent, and mitigate these threats. After analyzing the ways in which AI may influence the threat landscape in the digital, physical, and political domains, we make four high-level recommendations for AI researchers and other stakeholders. We also suggest several promising areas for further research that could expand the portfolio of defenses, or make attacks less effective or harder to execute. Finally, we discuss, but do not conclusively resolve, the long-term equilibrium of attackers and defenders.
We investigate evaluation metrics for dialogue response generation systems where supervised labels, such as task completion, are not available. Recent works in response generation have adopted metrics from machine translation to compare a model's generated response to a single target response. We show that these metrics correlate very weakly with human judgements in the non-technical Twitter domain, and not at all in the technical Ubuntu domain. We provide quantitative and qualitative results highlighting specific weaknesses in existing metrics, and provide recommendations for future development of better automatic evaluation metrics for dialogue systems.
Sentiment prediction of contemporary music can have a wide-range of applications in modern society, for instance, selecting music for public institutions such as hospitals or restaurants to potentially improve the emotional well-being of personnel, patients, and customers, respectively. In this project, music recommendation system built upon on a naive Bayes classifier, trained to predict the sentiment of songs based on song lyrics alone. The experimental results show that music corresponding to a happy mood can be detected with high precision based on text features obtained from song lyrics.
In the past decade, a lot of progress has been made in the design and evaluation of logic locking; a premier technique to safeguard the integrity of integrated circuits throughout the electronics supply chain. However, the widespread proliferation of machine learning has recently introduced a new pathway to evaluating logic locking schemes. This paper summarizes the recent developments in logic locking attacks and countermeasures at the frontiers of contemporary machine learning models. Based on the presented work, the key takeaways, opportunities, and challenges are highlighted to offer recommendations for the design of next-generation logic locking.
To provide AI researchers with modern tools for dealing with the explosive growth of the research literature in their field, we introduce a new platform, AI Research Navigator, that combines classical keyword search with neural retrieval to discover and organize relevant literature. The system provides search at multiple levels of textual granularity, from sentences to aggregations across documents, both in natural language and through navigation in a domain-specific Knowledge Graph. We give an overview of the overall architecture of the system and of the components for document analysis, question answering, search, analytics, expert search, and recommendations.
The Mapper algorithm does not include a check for whether the cover produced conforms to the requirements of the nerve lemma. To perform a check for obstructions to the nerve lemma, statistical considerations of multiple testing quickly arise. In this paper, we propose several statistical approaches to finding obstructions: through a persistent nerve lemma, through simulation testing, and using a parametric refinement of simulation tests. We suggest Certified Mapper -- a method built from these approaches to generate certificates of non-obstruction, or identify specific obstructions to the nerve lemma -- and we give recommendations for which statistical approaches are most appropriate for the task.
We present Etymo (https://etymo.io), a discovery engine to facilitate artificial intelligence (AI) research and development. It aims to help readers navigate a large number of AI-related papers published every week by using a novel form of search that finds relevant papers and displays related papers in a graphical interface. Etymo constructs and maintains an adaptive similarity-based network of research papers as an all-purpose knowledge graph for ranking, recommendation, and visualisation. The network is constantly evolving and can learn from user feedback to adjust itself.
We propose a new learning to rank algorithm, named Weighted Margin-Rank Batch loss (WMRB), to extend the popular Weighted Approximate-Rank Pairwise loss (WARP). WMRB uses a new rank estimator and an efficient batch training algorithm. The approach allows more accurate item rank approximation and explicit utilization of parallel computation to accelerate training. In three item recommendation tasks, WMRB consistently outperforms WARP and other baselines. Moreover, WMRB shows clear time efficiency advantages as data scale increases.
Controlled natural languages for industrial application are often regarded as a response to the challenges of translation and multilingual communication. This paper presents a quite different approach taken by Koenig & Bauer AG, where the main goal was the improvement of the authoring process for technical documentation. Most importantly, this paper explores the notion of a controlled language and demonstrates how style guides can emerge from non-linguistic considerations. Moreover, it shows the transition from loose language recommendations into precise and prescriptive rules and investigates whether such rules can be regarded as a full-fledged controlled language.