Search

Recent advances in multi-agent reinforcement learning (MARL) have created opportunities to solve complex real-world tasks. Cybersecurity is a notable application area, where defending networks against sophisticated adversaries remains a challenging task typically performed by teams of security operators. In this work, we explore novel MARL strategies for building autonomous cyber network defenses that address challenges such as large policy spaces, partial observability, and stealthy, deceptive adversarial strategies. To facilitate efficient and generalized learning, we propose a hierarchical Proximal Policy Optimization (PPO) architecture that decomposes the cyber defense task into specific sub-tasks like network investigation and host recovery. Our approach involves training sub-policies for each sub-task using PPO enhanced with domain expertise. These sub-policies are then leveraged by a master defense policy that coordinates their selection to solve complex network defense tasks. Furthermore, the sub-policies can be fine-tuned and transferred with minimal cost to defend against shifts in adversarial behavior or changes in network settings. We conduct extensive experiments using CybORG Cage 4, the state-of-the-art MARL environment for cyber defense. Comparisons with multiple baselines across different adversaries show that our hierarchical learning approach achieves top performance in terms of convergence speed, episodic return, and several interpretable metrics relevant to cybersecurity, including the fraction of clean machines on the network, precision, and false positives on recoveries., Comment: 9 pages, 7 figures, AAMAS preprint

Many molecular alterations serve as clinically prognostic or therapy-predictive biomarkers, typically detected using single or multi-gene molecular assays. However, these assays are expensive, tissue destructive and often take weeks to complete. Using AI on routine H&E WSIs offers a fast and economical approach to screen for multiple molecular biomarkers. We present a high-throughput AI-based system leveraging Virchow2, a foundation model pre-trained on 3 million slides, to interrogate genomic features previously determined by an next-generation sequencing (NGS) assay, using 47,960 scanned hematoxylin and eosin (H&E) whole slide images (WSIs) from 38,984 cancer patients. Unlike traditional methods that train individual models for each biomarker or cancer type, our system employs a unified model to simultaneously predict a wide range of clinically relevant molecular biomarkers across cancer types. By training the network to replicate the MSK-IMPACT targeted biomarker panel of 505 genes, it identified 80 high performing biomarkers with a mean AU-ROC of 0.89 in 15 most common cancer types. In addition, 40 biomarkers demonstrated strong associations with specific cancer histologic subtypes. Furthermore, 58 biomarkers were associated with targets frequently assayed clinically for therapy selection and response prediction. The model can also predict the activity of five canonical signaling pathways, identify defects in DNA repair mechanisms, and predict genomic instability measured by tumor mutation burden, microsatellite instability (MSI), and chromosomal instability (CIN). The proposed model can offer potential to guide therapy selection, improve treatment efficacy, accelerate patient screening for clinical trials and provoke the interrogation of new therapeutic targets.

The stability of geotechnical infrastructure assets, such as cuttings and embankments, is crucial to the safe and efficient delivery of transport services. The successful emulation of geotechnical models of deterioration of infrastructure slopes has the potential to inform slope design, maintenance and remediation by introducing the time dependency of deterioration into geotechnical asset management. We have performed computer experiments of deterioration, measured by the factor of safety (FoS), for a set of cutting slope geometries and soil properties that are common in the southern UK. Whilst computer experiments are an extremely useful and cost-effective method of better understanding deterioration mechanisms, it would not be practical to run enough experiments to understand relations between high-dimensional inputs and outputs. Therefore, we trained a fully-Bayesian Gaussian process emulator using an ensemble of 75 computer experiments to predict the FoS. We construct two different emulator models, one approximating the FoS temporal evolution with a quadratic model and one approximating the temporal evolution with a B-spline model; and we emulated their parameters. We also compare the ability of our models to predict failure time. The developed models could be used to inform infrastructure cutting slope design and management, and extend serviceable life.

Incorporating extra-textual context such as film metadata into the machine translation (MT) pipeline can enhance translation quality, as indicated by automatic evaluation in recent work. However, the positive impact of such systems in industry remains unproven. We report on an industrial case study carried out to investigate the benefit of MT in a professional scenario of translating TV subtitles with a focus on how leveraging extra-textual context impacts post-editing. We found that post-editors marked significantly fewer context-related errors when correcting the outputs of MTCue, the context-aware model, as opposed to non-contextual models. We also present the results of a survey of the employed post-editors, which highlights contextual inadequacy as a significant gap consistently observed in MT. Our findings strengthen the motivation for further work within fully contextual MT., Comment: Accepted to EAMT 2024

The resolvent degree $\textrm{rd}_{\mathbb{C}}(n)$ is the smallest integer $d$ such that a root of the general polynomial $$f(x) = x^n + a_1 x^{n-1} + \ldots + a_n$$ can be expressed as a composition of algebraic functions in at most $d$ variables with complex coefficients. It is known that $\textrm{rd}_{\mathbb{C}}(n) = 1$ when $n \leqslant 5$. Hilbert was particularly interested in the next three cases: he asked if $\textrm{rd}_{\mathbb{C}}(6) = 2$ (Hilbert's Sextic Conjecture), $\textrm{rd}_{\mathbb{C}}(7) = 3$ (Hilbert's 13th Problem) and $\textrm{rd}_{\mathbb{C}}(8) = 4$ (Hilbert's Octic Conjecture). These problems remain open. It is known that $\textrm{rd}_{\mathbb{C}}(6) \leqslant 2$, $\textrm{rd}_{\mathbb{C}}(7) \leqslant 3$ and $\textrm{rd}_{\mathbb{C}}(8) \leqslant 4$. It is not known whether or not $\textrm{rd}_{\mathbb{C}}(n)$ can be $> 1$ for any $n \geqslant 6$. In this paper, we show that all three of Hilbert's conjectures can fail if we replace $\mathbb C$ with a base field of positive characteristic., Comment: 12 pages, 2 figures

Group dance, a sub-genre characterized by intricate motions made by a cohort of performers in tight synchronization, has a longstanding and culturally significant history and, in modern forms such as cheerleading, a broad base of current adherents. However, despite its popularity, learning group dance routines remains challenging. Based on the prior success of interactive systems to support individual dance learning, this paper argues that group dance settings are fertile ground for augmentation by interactive aids. To better understand these design opportunities, this paper presents a sequence of user-centered studies of and with amateur cheerleading troupes, spanning from the formative (interviews, observations) through the generative (an ideation workshop) to concept validation (technology probes and speed dating). The outcomes are a nuanced understanding of the lived practice of group dance learning, a set of interactive concepts to support those practices, and design directions derived from validating the proposed concepts. Through this empirical work, we expand the design space of interactive dance practice systems from the established context of single-user practice (primarily focused on gesture recognition) to a multi-user, group-based scenario focused on feedback and communication., Comment: 20 pages, 10 figures, 1 table, to be published in the proceedings of the ACM Designing Interactive Systems Conference, 2024, (DIS '24)

Foundation models in computational pathology promise to unlock the development of new clinical decision support systems and models for precision medicine. However, there is a mismatch between most clinical analysis, which is defined at the level of one or more whole slide images, and foundation models to date, which process the thousands of image tiles contained in a whole slide image separately. The requirement to train a network to aggregate information across a large number of tiles in multiple whole slide images limits these models' impact. In this work, we present a slide-level foundation model for H&E-stained histopathology, PRISM, that builds on Virchow tile embeddings and leverages clinical report text for pre-training. Using the tile embeddings, PRISM produces slide-level embeddings with the ability to generate clinical reports, resulting in several modes of use. Using text prompts, PRISM achieves zero-shot cancer detection and sub-typing performance approaching and surpassing that of a supervised aggregator model. Using the slide embeddings with linear classifiers, PRISM surpasses supervised aggregator models. Furthermore, we demonstrate that fine-tuning of the PRISM slide encoder yields label-efficient training for biomarker prediction, a task that typically suffers from low availability of training data; an aggregator initialized with PRISM and trained on as little as 10% of the training data can outperform a supervised baseline that uses all of the data.

Recent breakthroughs in large models have highlighted the critical significance of data scale, labels and modals. In this paper, we introduce MS MARCO Web Search, the first large-scale information-rich web dataset, featuring millions of real clicked query-document labels. This dataset closely mimics real-world web document and query distribution, provides rich information for various kinds of downstream tasks and encourages research in various areas, such as generic end-to-end neural indexer models, generic embedding models, and next generation information access system with large language models. MS MARCO Web Search offers a retrieval benchmark with three web retrieval challenge tasks that demand innovations in both machine learning and information retrieval system research domains. As the first dataset that meets large, real and rich data requirements, MS MARCO Web Search paves the way for future advancements in AI and system research. MS MARCO Web Search dataset is available at: https://github.com/microsoft/MS-MARCO-Web-Search., Comment: 10 pages, 6 figures, for associated dataset, see http://github.com/microsoft/MS-MARCO-Web-Search

The adoption of the Industrial Internet of Things (IIoT) as a complementary technology to Operational Technology (OT) has enabled a new level of standardised data access and process visibility. This convergence of Information Technology (IT), OT, and IIoT has also created new cybersecurity vulnerabilities and risks that must be managed. Artificial Intelligence (AI) is emerging as a powerful tool to monitor OT/IIoT networks for malicious activity and is a highly active area of research. AI researchers are applying advanced Machine Learning (ML) and Deep Learning (DL) techniques to the detection of anomalous or malicious activity in network traffic. They typically use datasets derived from IoT/IIoT/OT network traffic captures to measure the performance of their proposed approaches. Therefore, there is a widespread need for datasets for algorithm testing. This work systematically reviews publicly available network traffic capture-based datasets, including categorisation of contained attack types, review of metadata, and statistical as well as complexity analysis. Each dataset is analysed to provide researchers with metadata that can be used to select the best dataset for their research question. This results in an added benefit to the community as researchers can select the best dataset for their research more easily and according to their specific Machine Learning goals., Comment: 28 pages, preprint submitted to Engineering Applications of Artificial Intelligence (Elsevier)

Serverless computing offers attractive scalability, elasticity and cost-effectiveness. However, constraints on memory, CPU and function runtime have hindered its adoption for data-intensive applications and machine learning (ML) workloads. Traditional 'server-ful' platforms enable distributed computation via fast networks and well-established inter-process communication (IPC) mechanisms such as MPI and shared memory. In the absence of such solutions in the serverless domain, parallel computation with significant IPC requirements is challenging. We present FSD-Inference, the first fully serverless and highly scalable system for distributed ML inference. We explore potential communication channels, in conjunction with Function-as-a-Service (FaaS) compute, to design a state-of-the-art solution for distributed ML within the context of serverless data-intensive computing. We introduce novel fully serverless communication schemes for ML inference workloads, leveraging both cloud-based publish-subscribe/queueing and object storage offerings. We demonstrate how publish-subscribe/queueing services can be adapted for FaaS IPC with comparable performance to object storage, while offering significantly reduced cost at high parallelism levels. We conduct in-depth experiments on benchmark DNNs of various sizes. The results show that when compared to server-based alternatives, FSD-Inference is significantly more cost-effective and scalable, and can even achieve competitive performance against optimized HPC solutions. Experiments also confirm that our serverless solution can handle large distributed workloads and leverage high degrees of FaaS parallelism., Comment: In Proceedings of 2024 IEEE 40th International Conference on Data Engineering (ICDE) (to appear)

Programmatically generating tight differential privacy (DP) bounds is a hard problem. Two core challenges are (1) finding expressive, compact, and efficient encodings of the distributions of DP algorithms, and (2) state space explosion stemming from the multiple quantifiers and relational properties of the DP definition. We address the first challenge by developing a method for tight privacy and accuracy bound synthesis using weighted model counting on binary decision diagrams, a state-of-the-art technique from the artificial intelligence and automated reasoning communities for exactly computing probability distributions. We address the second challenge by developing a framework for leveraging inherent symmetries in DP algorithms. Our solution benefits from ongoing research in probabilistic programming languages, allowing us to succinctly and expressively represent different DP algorithms with approachable language syntax that can be used by non-experts. We provide a detailed case study of our solution on the binary randomized response algorithm. We also evaluate an implementation of our solution using the Dice probabilistic programming language for the randomized response and truncated geometric above threshold algorithms. We compare to prior work on exact DP verification using Markov chain probabilistic model checking and the decision procedure DiPC. Very few existing works consider mechanized analysis of accuracy guarantees for DP algorithms. We additionally provide a detailed analysis using our technique for finding tight accuracy bounds for DP algorithms., Comment: In IEEE 37th Computer Security Foundations Symposium (CSF) 2024

An assurance calculation is a Bayesian alternative to a power calculation. One may be performed to aid the planning of a clinical trial, specifically setting the sample size or to support decisions about whether or not to perform a study. Immuno-oncology is a rapidly evolving area in the development of anticancer drugs. A common phenomenon that arises in trials of such drugs is one of delayed treatment effects, that is, there is a delay in the separation of the survival curves. To calculate assurance for a trial in which a delayed treatment effect is likely to be present, uncertainty about key parameters needs to be considered. If uncertainty is not considered, the number of patients recruited may not be enough to ensure we have adequate statistical power to detect a clinically relevant treatment effect and the risk of an unsuccessful trial is increased. We present a new elicitation technique for when a delayed treatment effect is likely and show how to compute assurance using these elicited prior distributions. We provide an example to illustrate how this can be used in practice and develop open-source software to implement our methods. Our methodology has the potential to improve the success rate and efficiency of Phase III trials in immuno-oncology and for other treatments where a delayed treatment effect is expected to occur.

The use of artificial intelligence to enable precision medicine and decision support systems through the analysis of pathology images has the potential to revolutionize the diagnosis and treatment of cancer. Such applications will depend on models' abilities to capture the diverse patterns observed in pathology images. To address this challenge, we present Virchow, a foundation model for computational pathology. Using self-supervised learning empowered by the DINOv2 algorithm, Virchow is a vision transformer model with 632 million parameters trained on 1.5 million hematoxylin and eosin stained whole slide images from diverse tissue and specimen types, which is orders of magnitude more data than previous works. The Virchow model enables the development of a pan-cancer detection system with 0.949 overall specimen-level AUC across 17 different cancer types, while also achieving 0.937 AUC on 7 rare cancer types. The Virchow model sets the state-of-the-art on the internal and external image tile level benchmarks and slide level biomarker prediction tasks. The gains in performance highlight the importance of training on massive pathology image datasets, suggesting scaling up the data and network architecture can improve the accuracy for many high-impact computational pathology applications where limited amounts of training data are available.

We prove that for hyperbolic fibered knots in any closed, connected, oriented 3-manifold the volume and genus are unrelated. As an application we answer a question of Hirose, Kalfagianni, and Kin about volumes of mapping tori that are double branched covers., Comment: 9 pages, 5 figures

The essential dimension $\operatorname{ed}_k({\rm S}_n)$ of the symmetric group ${\rm S}_n$ is the minimal integer $d$ such that the general polynomial $x^n + a_1 x^{n-1} + \ldots + a_n$ can be reduced to a $d$-parameter form by a Tschirnhaus transformation. Finding this number is a long-standing open problem, originating in the work of Felix Klein, long before essential dimension was formally defined. We now know that $\operatorname{ed}_k({\rm S}_n)$ lies between $\lfloor n/2 \rfloor$ and $n-3$ for every $n \geqslant 5$ and every field $k$ of characteristic different from $2$. Moreover, if $\operatorname{char}(k) = 0$, then $\operatorname{ed}_k({\rm S}_n) \geqslant \lfloor (n+1)/2 \rfloor$ for any $n \geqslant 6$. The value of $\operatorname{ed}_k({\rm S}_n)$ is not known for any $n \geqslant 8$ and any field $k$, though it is widely believed that $\operatorname{ed}_k({\rm S}_n)$ should be $n-3$ for every $n \geqslant 5$, at least in characteristic $0$. In this paper we show that for every odd prime $p$ there are infinitely many positive integers $n$ such that $\operatorname{ed}_{\mathbb F_p}(\rm{S}_n) \leqslant n-4$., Comment: 8 pages