AAIP Fellow Xingyu Zhao and Comp Sci, Liverpool colleagues will present this new paper on dependability analysis of deep reinforcement learning IROS 2022. Available on Arxiv now: arxiv.org/abs/2109.06523

A new joint Topic "#Safe and Secure #Autonomous Systems" is open for submission: mdpi.com/topics/Autonom…. This article collection aims to bring together research on how to identify, rectify, and verify the safety and #security risks of learning-enabled systems. #openaccess

Questions for a quick assessment of an ML interpretation method: • Are the explanations global or local? • Does the method interpret model components or model behavior? • Does the method compute feature effects, feature importance, or attributions?

At UoL, we have 2 openings in Computer Science at Lecturer level (= Assistant Prof), deadline 2022-11-08. Our overarching research themes are: - Algorithms and Networks - Artificial Intelligence - Economics and Computation - Logical Methods in CS jobs.ac.uk/job/CUA711/lec…

Morning! First time visit the new building @AAIP_York

New paper from the #AssuringAutonomy community. Programme Fellows Xingyu Zhao & Al Banks Dstl are two authors on this paper that presents an assurance framework for learning-enabled systems. Preprint available: arxiv.org/abs/2112.00646

New paper from #AssuringAutonomy Fellow Xingyu Zhao and colleague Kizito Salako. Accepted by IEEE Transactions on Software Engineering, the Arxiv version is available now: arxiv.org/pdf/2208.00462…

Our multidisciplinary community is influencing change. Chapter 2 of the #AssuringAutonomy Impact Report expresses the significance of the AAIP community and our collaborators in tackling complex issues and leading to new research outcomes. Read the report: bit.ly/41hxA8E

Our team (“University of Liverpool”), one of the winners of Phase 1 of the UK-US PETs Prize Challenges, is honoured to receive a special recognition “Novel Modelling/Design” prize in the Phase 2 of UK-US PETs Prize Challenges for our solution which includ…lnkd.in/euyECNhF

:) fun time with ChatGPT, and learn how they might help us on safety analysis. lnkd.in/eTEhBpXj

1. Reminder of what ‘net zero’ really means. This graphic from UK Govt FIRES project. Key points: all airports except Heathrow, Belfast & Glasgow to close by 2030. NO FLYING at all by 2050. No new petrol/diesel cars by 2030; by 2050 road use restricted to 60% of today’s level.

#WAISE 2023 is back for its 6th edition! We look forward to your contributions on everything related to AI Safety Engineering 🦺🤖🚕🧠 🗓️ Submission: May 5th; Notification: May 31st waise.org (RTs welcome)

Would be great if we can receive submissions regarding large language models and ChatGPT for interesting discussions in the workshop. lnkd.in/eMcP4qB5

LLMs -- we completed a survey with 300+ references, trying to summarise the known vulnerabilities of LLMs and discuss whether and how the verification and validation (V&V) techniques can be adapted to work with LLMs. The paper is now available at ArXiv: lnkd.in/eK5EEQKS .

We have always paved new roads with our unique end-to-end AI approach. Today, we are excited to announce our collaboration with WMG on the CCAV, @InnovateUK backed project: #DriveSafeAI to support the safety assurance of AI-based self-driving tech wayve.ai/press/drivesaf…

Every Data Scientist needs to know these ideas. They will blow your mind. 1. Correlation vs Causation P(A | B) is the probability of A given B. It is the probability that we will observe A given that we have already observed B. P(A | do(B)) is the probability of A given

Oh no... 🤣 While this is a good "option", I guess this will make some (most?) authors' life harder. As anticipated, we will see many "whole revised manuscripts". Having constraints can be beneficial, as opposed to open another racing track. #icml2024

#WAISE24 is back for its 7th edition 🤖🦺🪖 We look forward to your contributions on topics related to Safe AI and AI for Safety Engineering, Foundational models, LLMs 🔥 etc. 🌐waise.org 🗓️ Submission: May 5 🗓️ Notification: May 31 🗓️ Event: Sept 17 (RTs welcome)

And the ACM A.M. Turing Award goes to... bit.ly/3UeG13a #TuringAward