DOCSLIB.ORG

ICML 2019 Workshop Book Generated Sat Jul 06, 2019

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
ICML 2019 Workshop Book Generated Sat Jul 06, 2019 ICML 2019 Workshop book Generated Sat Jul 06, 2019 Workshop organizers make last-minute changes to 204, Negative Dependence: Theory and their schedule. Download this document again to Applications in Machine Learning Gartrell, get the lastest changes, or use the ICML mobile Gillenwater, Kulesza, Mariet application. Grand Ballroom A, Understanding and Improving Schedule Highlights Generalization in Deep Learning Krishnan, Mobahi, Neyshabur, Bartlett, Song, Srebro June 14, 2019 Grand Ballroom B, 6th ICML Workshop on 101, ICML 2019 Workshop on Computational Automated Machine Learning (AutoML 2019) Biology Pe'er, Prabhakaran, Azizi, Diallo, Hutter, Vanschoren, Eggensperger, Feurer Kundaje, Engelhardt, Dhifli, MEPHU NGUIFO, Hall A, Generative Modeling and Model-Based Tansey, Vogt, Listgarten, Burdziak, CompBio Reasoning for Robotics and AI Rajeswaran, 102, ICML 2019 Time Series Workshop Todorov, Mordatch, Agnew, Zhang, Pineau, Kuznetsov, Yang, Yu, Tang, Wang Chang, Erhan, Levine, Stachenfeld, Zhang 103, Human In the Loop Learning (HILL) Wang, Hall B, Uncertainty and Robustness in Deep Wang, Yu, Zhang, Gonzalez, Jia, Bird, Learning Li, Lakshminarayanan, Hendrycks, Varshney, Kim, Weller Dietterich, Gilmer 104 A, Climate Change: How Can AI Help? Seaside Ballroom, Reinforcement Learning for Rolnick, Lacoste, Maharaj, Chayes, Bengio Real Life Li, Geramifard, Li, Szepesvari, Wang 104 B, Workshop on the Security and Privacy of Seaside Ballroom, Real-world Sequential Machine Learning Papernot, Tramer, Li, Decision Making: Reinforcement Learning Boneh, Evans, Jha, Liang, McDaniel, Steinhardt, and Beyond Le, Yue, Swaminathan, Boots, Song Cheng 104 C, Theoretical Physics for Deep Learning June 15, 2019 Lee, Pennington, Bahri, Welling, Ganguli, Bruna 101, Workshop on AI for autonomous driving 201, AI in Finance: Applications and Choromanska, Jackel, Li, Niebles, Gaidon, Infrastructure for Multi-Agent Learning Posner, Chao Reddy, Balch, Wellman, Kumar, Stoica, Elkind 102, Workshop on Multi-Task and Lifelong 202, The Third Workshop On Tractable Reinforcement Learning Chandar, Sodhani, Probabilistic Modeling (TPM) Lowd, Vergari, Khetarpal, Zahavy, Mankowitz, Mannor, Molina, Rahman, Domingos, Vergari Ravindran, Precup, Finn, Gupta, Zhang, Cho, Rusu, Rob Fergus 203, Joint Workshop on On-Device Machine Learning & Compact Deep Neural Network 103, Invertible Neural Networks and Normalizing Representations (ODML-CDNNR) Ravi, Flows Huang, Krueger, Van den Berg, Kozareva, Fan, Welling, Chen, Bailer, Kulis, Hu, Papamakarios, Gomez, Cremer, Chen, Dekhtiar, Lin, Marculescu Courville, J. Rezende Page 1 of 107 ICML 2019 Workshop book Generated Sat Jul 06, 2019 104 A, Stein’s Method for Machine Learning and Seaside Ballroom, Adaptive and Multitask Statistics Briol, Mackey, Oates, Liu, Goldstein Learning: Algorithms & Systems Al-Shedivat, Platanios, Stretcu, Andreas, Talwalkar, 104 B, AI For Social Good (AISG) Luck, Sankaran, Caruana, Mitchell, Xing Sylvain, McGregor, Penn, Tadesse, Sylvain, Côté, Mackey, Ghani, Bengio 104 C, Synthetic Realities: Deep Learning for Detecting AudioVisual Fakes Biggio, Korshunov, Mensink, Patrini, Sadhu, Rao 201, ICML Workshop on Imitation, Intent, and Interaction (I3) Rhinehart, Levine, Finn, He, Kostrikov, Fu, Reddy 202, Coding Theory For Large-scale Machine Learning Cadambe, Grover, Papailiopoulos, Joshi 203, The How2 Challenge: New Tasks for Vision & Language Metze, Specia, Elliot, Barrault, Sanabria, Palaskar 204, Machine Learning for Music Discovery Schmidt, Nieto, Gouyon, Kinnaird, Lanckriet Grand Ballroom A, Workshop on Self-Supervised Learning van den Oord, Aytar, Doersch, Vondrick, Radford, Sermanet, Zamir, Abbeel Grand Ballroom B, Learning and Reasoning with Graph-Structured Representations Fetaya, Hu, Kipf, Li, Liang, Liao, Urtasun, Wang, Welling, Xing, Zemel Hall A, Exploration in Reinforcement Learning Workshop Bhupatiraju, Eysenbach, Gu, Edwards, White, Oudeyer, Stanley, Brunskill Hall B, Identifying and Understanding Deep Learning Phenomena Sedghi, Bengio, Hata, Madry, Morcos, Neyshabur, Raghu, Rahimi, Schmidt, Xiao Page 2 of 107 ICML 2019 Workshop book Generated Sat Jul 06, 2019 Bioinformatics and AI at IJCAI 2015 Buenos Aires, June 14, 2019 IJCAI 2016 New York, IJCAI 2017 Melbourne which had excellent line-ups of talks and were well-received by the community. Every year, we received 60+ submissions. After multiple rounds of rigorous reviewing around 50 submissions were selected from which the best set of papers were ICML 2019 Workshop on Computational chosen for Contributed talks and Spotlights and the Biology rest were invited as Posters. We have a steadfast and growing base of reviewers making up the Donna Pe'er, Sandhya Prabhakaran, Elham Program Committee. For the past edition, a special Azizi, Abdoulaye Baniré Diallo, Anshul Kundaje, issue of Journal of Computational Biology will be Barbara Engelhardt, Wajdi Dhifli, Engelbert released in the following weeks with extended MEPHU NGUIFO, Wesley Tansey, Julia Vogt, versions of 14 accepted papers. Jennifer Listgarten, Cassandra Burdziak, Workshop CompBio We have two confirmed invited speakers and we will invite at least one more leading researcher in the 101, Fri Jun 14, 08:30 AM field. Similar to previous years, we plan to request partial funding from Microsoft Research, Google, The workshop will showcase recent research in the Python, Swiss Institute of Bioinformatics that we field of Computational Biology. There has been intend to use for student travel awards. In past significant development in genomic sequencing years, we have also been able to provide awards for techniques and imaging technologies. These best poster/paper and partially contribute to travel approaches not only generate huge amounts of expenses for at least 8 students per year. data but provide unprecedented resolution of single cells and even subcellular structures. The The Workshop proceedings will be available availability of high dimensional data, at multiple through CEUR proceedings. We would also have an spatial and temporal resolutions has made machine extended version to be included in a special issue of learning and deep learning methods increasingly the Journal of Computational Biology (JCB) for critical for computational analysis and interpretation which we already have an agreement with JCB. of the data. Conversely, biological data has also exposed unique challenges and problems that call Schedule for the development of new machine learning methods. This workshop aims to bring together 08:30 researchers working at the intersection of Machine WCB Organisers AM Learning and Biology to present recent advances 08:40 and open questions in Computational Biology to the Caroline Uhler ML community. AM Jacopo Cirrone, The workshop is a sequel to the WCB workshops Marttinen Pekka , 09:20 we organized in the last three years Joint ICML and Elior Rahmani, Elior AM IJCAI 2018, Stockholm, ICML 2017, Sydney and Rahmani, Harri ICML 2016, New York as well as Workshop on Lähdesmäki Page 3 of 107 ICML 2019 Workshop book Generated Sat Jul 06, 2019 09:50 ICML 2019 Time Series Workshop Ali Oskooei AM Vitaly Kuznetsov, Scott Yang, Rose Yu, Cheng 11:00 Francisco LePort Tang, Yuyang Wang AM 102, Fri Jun 14, 08:30 AM Ali Oskooei, 11:40 Zhenqin Wu, Karren Time series data is both quickly growing and AM Dai Yang, Mijung already ubiquitous. In domains spanning as broad a Kim range as climate, robotics, entertainment, finance, Wiese, Carter, healthcare, and transportation, there has been a DeBlasio, Hashir, significant shift away from parsimonious, infrequent Chan, Manica, measurements to nearly continuous monitoring and Oskooei, Wu, Yang, recording. Rapid advances in sensing technologies, FAGES, Liu, ranging from remote sensors to wearables and Beebe-Wang, He, social sensing, are generating rapid growth in the 12:00 Poster Session & Cirrone, Marttinen, size and complexity of time series data streams. PM Lunch break Rahmani, Thus, the importance and impact of time series Lähdesmäki, Yadala, analysis and modelling techniques only continues to Deac, Soleimany, grow. Mane, Ernst, Cohen, Mathew, Agarwal, At the same time, while time series analysis has Zheng been extensively studied by econometricians and 02:00 statisticians, modern time series data often pose Ngo Trong Trung PM significant challenges for the existing techniques both in terms of their structure (e.g., irregular Achille Nazaret, sampling in hospital records and spatiotemporal François Fages, structure in climate data) and size. Moreover, the 02:20 Brandon Carter, focus on time series in the machine learning PM Ruishan Liu, community has been comparatively much smaller. Nicasia In fact, the predominant methods in machine Beebe-Wang learning often assume i.i.d. data streams, which is 02:50 Poster Session & generally not appropriate for time series data. Thus, PM Coffee Break there is both a great need and an exciting opportunity for the machine learning community to 04:00 Daphne Koller develop theory, models and algorithms specifically PM for the purpose of processing and analyzing time 04:40 series data. Bryan He PM 05:00 Award Ceremony & We see ICML as a great opportunity to bring PM Closing Remarks together theoretical and applied researchers from around the world and with different backgrounds who are interested in the development and usage of time series analysis and algorithms. This includes Page 4 of 107 ICML 2019 Workshop book Generated Sat Jul 06, 2019 methods for time series prediction, classification, workshop includes people who
Load more

Recommended publications
  • Professor Marta Kwiatkowska
    Potential Supervisors Marta Kwiatkowska Marta Kwiatkowska is Professor of Computing Systems and Fellow of Trinity College, University of Oxford. Prior to this she was Professor in the School of Computer Science at the University of Birmingham, Lecturer at the University of Leicester and Assistant Professor at the Jagiellonian University in Cracow, Poland. She holds a BSc/MSc in Computer Science from the Jagiellonian University, MA from Oxford and a PhD from the University of Leicester. In 2014 she was awarded an honorary doctorate from KTH Royal Institute of Technology in Stockholm. Marta Kwiatkowska spearheaded the development of probabilistic and quantitative methods in verification on the international scene. She led the development of the PRISM model checker, the leading software tool in the area and widely used for research and teaching and winner of the HVC 2016 Award. Applications of probabilistic model checking have spanned communication and security protocols, nanotechnology designs, power management, game theory, planning and systems biology, with genuine flaws found and corrected in real-world protocols. Kwiatkowska gave the Milner Lecture in 2012 in recognition of "excellent and original theoretical work which has a perceived significance for practical computing" and was invited to give keynotes at the LICS 2003, ESEC/FSE 2007, ETAPS/FASE 2011, ATVA 2013, ICALP 2016 and CAV 2017 conferences. Marta Kwiatkowska is the first female winner of the 2018 Royal Society Milner Award and Lecture. She is a Fellow of ACM, member of Academia Europea, Fellow of EATCS and Fellow of the BCS. She serves on editorial boards of several journals, including Information and Computation, Formal Methods in System Design, Logical Methods in Computer Science, Science of Computer Programming and Royal Society Open Science journal.
    [Show full text]
  • Current Issue of FACS FACTS
    Issue 2021-2 July 2021 FACS A C T S The Newsletter of the Formal Aspects of Computing Science (FACS) Specialist Group ISSN 0950-1231 FACS FACTS Issue 2021-2 July 2021 About FACS FACTS FACS FACTS (ISSN: 0950-1231) is the newsletter of the BCS Specialist Group on Formal Aspects of Computing Science (FACS). FACS FACTS is distributed in electronic form to all FACS members. Submissions to FACS FACTS are always welcome. Please visit the newsletter area of the BCS FACS website for further details at: https://www.bcs.org/membership/member-communities/facs-formal-aspects- of-computing-science-group/newsletters/ Back issues of FACS FACTS are available for download from: https://www.bcs.org/membership/member-communities/facs-formal-aspects- of-computing-science-group/newsletters/back-issues-of-facs-facts/ The FACS FACTS Team Newsletter Editors Tim Denvir [email protected] Brian Monahan [email protected] Editorial Team: Jonathan Bowen, John Cooke, Tim Denvir, Brian Monahan, Margaret West. Contributors to this issue: Jonathan Bowen, Andrew Johnstone, Keith Lines, Brian Monahan, John Tucker, Glynn Winskel BCS-FACS websites BCS: http://www.bcs-facs.org LinkedIn: https://www.linkedin.com/groups/2427579/ Facebook: http://www.facebook.com/pages/BCS-FACS/120243984688255 Wikipedia: http://en.wikipedia.org/wiki/BCS-FACS If you have any questions about BCS-FACS, please send these to Jonathan Bowen at [email protected]. 2 FACS FACTS Issue 2021-2 July 2021 Editorial Dear readers, Welcome to the 2021-2 issue of the FACS FACTS Newsletter. A theme for this issue is suggested by the thought that it is just over 50 years since the birth of Domain Theory1.
    [Show full text]
  • Dr Gordon Baxter, University of St Andrews
    Dr Gordon Baxter, University of St Andrews Gordon Baxter has degrees in Computer Science (BSc(Eng)), Cognitive Science (MPhil), and Human Factors (PhD). He has several years experience in systems development, mostly on industrial and defence applications, having worked for Rolls Royce, Racal, Systems Designers, NEI and ICI. Most of this work involved aspects related to user interface design and human-computer interaction. Since joining academia he has worked on projects looking at rapid decision making in complex environments (for the DRA), dependability of computer based systems (for the NHS in Leeds, and several social services departments across the UK), and large scale complex IT systems. He is a Chartered Engineer, a Member of the British Computer Society, and an Associate Member of the Ergonomics Society. Dr Radu Calinescu, University of Oxford Radu Calinescu is a senior research officer at Oxford University Computing Laboratory. He obtained his DPhil in Computation from the University of Oxford in 1998, with a thesis that received a Distinguished Dissertation Award from the British Computer Society and was published as a Springer book. From 1999 to 2005, he worked in industry as the technical architect and senior development manager of an Oxford University spin-out software company specialising in the development of policy-driven data-centre management solutions. His research interests include autonomic computing, policy-based management of complex systems, modelling and model checking, automated and model-driven software engineering, and high performance computing. He is a senior member of the IEEE, and has chaired or served on the technical program committees of several international conferences on autonomic computing and formal methods.
    [Show full text]
  • Mathematics People
    NEWS Mathematics People or up to ten years post-PhD, are eligible. Awardees receive Braverman Receives US$1 million distributed over five years. NSF Waterman Award —From an NSF announcement Mark Braverman of Princeton University has been selected as a Prizes of the Association cowinner of the 2019 Alan T. Wa- terman Award of the National Sci- for Women in Mathematics ence Foundation (NSF) for his work in complexity theory, algorithms, The Association for Women in Mathematics (AWM) has and the limits of what is possible awarded a number of prizes in 2019. computationally. According to the Catherine Sulem of the Univer- prize citation, his work “focuses on sity of Toronto has been named the Mark Braverman complexity, including looking at Sonia Kovalevsky Lecturer for 2019 by algorithms for optimization, which, the Association for Women in Math- when applied, might mean planning a route—how to get ematics (AWM) and the Society for from point A to point B in the most efficient way possible. Industrial and Applied Mathematics “Algorithms are everywhere. Most people know that (SIAM). The citation states: “Sulem every time someone uses a computer, algorithms are at is a prominent applied mathemati- work. But they also occur in nature. Braverman examines cian working in the area of nonlin- randomness in the motion of objects, down to the erratic Catherine Sulem ear analysis and partial differential movement of particles in a fluid. equations. She has specialized on “His work is also tied to algorithms required for learning, the topic of singularity development in solutions of the which serve as building blocks to artificial intelligence, and nonlinear Schrödinger equation (NLS), on the problem of has even had implications for the foundations of quantum free surface water waves, and on Hamiltonian partial differ- computing.
    [Show full text]
  • Governing Machine Learning That Matters
    Doctoral thesis Governing Machine Learning that Matters Michael Veale 2019 A dissertation submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Science, Technology, Engineering and Public Policy Department of Science, Technology, Engineering and Public Policy (STEaPP) University College London 84,570 words 2019 Declaration of Authorship I, Michael Veale, confirm that the work presented in this thesis is my own. Wherein- formation has been derived from other sources, I confirm that this has been indicated in the thesis. Signed: Declaration of Integrated Publications This section acknowledges the integration of work of the Author into the different sec- tions of this thesis. All work integrated into the thesis was undertaken during the period in which the Author was registered with the University as working towards the doctorate. The work is acknowledged here as, since publication, other researchers have responded to and engaged with these works in this fast-moving field, and this thesis represents both a statement of the original arguments and findings in those works, as well as a partial response to the research field as it stands at the time of sub- mission. Chapter 1, Hello, World!, includes some content from the following articles: 1. Vasilios Mavroudis and Michael Veale, ‘Eavesdropping Whilst You’re Shopping: Balancing Personalisation and Privacy in Connected Retail Spaces’ in Proceedings of the 2018 PETRAS/IoTUK/IET Living in the IoT Conference (IET 2018) DOI: 10/gffng2; 2. Lilian Edwards and Michael Veale, ‘Slave to the Algorithm? Why a ‘Right to an Explanation’ is Probably Not The Remedy You Are Looking For’ (2017) 16 Duke L.
    [Show full text]
  • Safety Verification for Deep Neural
    1 Safety Verification for Deep Neural Networks with 2 Provable Guarantees 3 Marta Kwiatkowska 4 Department of Computer Science, University of Oxford, UK 5 http://http://www.cs.ox.ac.uk/marta.kwiatkowska/ 6 [email protected] 7 Abstract 8 Computing systems are becoming ever more complex, increasingly often incorporating deep learning 9 components. Since deep learning is unstable with respect to adversarial perturbations, there is a 10 need for rigorous software development methodologies that encompass machine learning. This paper 11 describes progress with developing automated verification techniques for deep neural networks to 12 ensure safety and robustness of their decisions with respect to input perturbations. This includes 13 novel algorithms based on feature-guided search, games, global optimisation and Bayesian methods. 14 2012 ACM Subject Classification Theory of computation → Logic and verification; Computing 15 methodologies → Neural networks 16 Keywords and phrases Neural networks, robustness, formal verification, Bayesian neural networks 17 Digital Object Identifier 10.4230/LIPIcs.CONCUR.2019.38 18 Category Invited paper 19 Funding Marta Kwiatkowska: This project has received funding from the European Research 20 Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant 21 agreement No. 834115) and the EPSRC Programme Grant on Mobile Autonomy (EP/M019918/1). 22 1 Introduction 23 Computing devices have become ubiquitous and ever present in our lives: smartphones 24 help us stay in touch with family and friends, GPS-enabled apps offer directions literally 25 at our fingertips, and voice-controlled assistants are now able to execute simple commands. 26 Artificial Intelligence is making great strides, promising many more exciting applications 27 with an increased level of autonomy, from wearable medical devices to robotic care assistants 28 and self-driving cars.
    [Show full text]
  • Arxiv:1906.09142V3 [Cs.LO] 17 Jul 2019 Reason About a Range of Quantitative Aspects of a System’S Behaviour: Probability, Time, Resource Usage, and Many Others
    Verification and Control of Turn-Based Probabilistic Real-Time Games Marta Kwiatkowska1, Gethin Norman2, and David Parker3 1 Department of Computing Science, University of Oxford, UK 2 School of Computing Science, University of Glasgow, UK 3 School of Computer Science, University of Birmingham, UK Abstract. Quantitative verification techniques have been developed for the formal analysis of a variety of probabilistic models, such as Markov chains, Markov decision process and their variants. They can be used to produce guarantees on quantitative aspects of system behaviour, for ex- ample safety, reliability and performance, or to help synthesise controllers that ensure such guarantees are met. We propose the model of turn-based probabilistic timed multi-player games, which incorporates probabilistic choice, real-time clocks and nondeterministic behaviour across multiple players. Building on the digital clocks approach for the simpler model of probabilistic timed automata, we show how to compute the key mea- sures that underlie quantitative verification, namely the probability and expected cumulative price to reach a target. We illustrate this on case studies from computer security and task scheduling. 1 Introduction Probability is a crucial tool for modelling computerised systems. We can use it to model uncertainty, for example in the operating environment of an autonomous vehicle or a wireless sensor network, and we can reason about systems that use randomisation, from probabilistic routing in anonymity network protocols to symmetry breaking in communication protocols. Formal verification of such systems can provide us with rigorous guarantees on, for example, the performance and reliability of computer networks [7], the amount of inadvertent information leakage by a security protocol [5], or the safety level of an airbag control system [2].
    [Show full text]
  • Michel Foucault Ronald C Kessler Graham Colditz Sigmund Freud
    ANK RESEARCHER ORGANIZATION H INDEX CITATIONS 1 Michel Foucault Collège de France 296 1026230 2 Ronald C Kessler Harvard University 289 392494 3 Graham Colditz Washington University in St Louis 288 316548 4 Sigmund Freud University of Vienna 284 552109 Brigham and Women's Hospital 5 284 332728 JoAnn E Manson Harvard Medical School 6 Shizuo Akira Osaka University 276 362588 Centre de Sociologie Européenne; 7 274 771039 Pierre Bourdieu Collège de France Massachusetts Institute of Technology 8 273 308874 Robert Langer MIT 9 Eric Lander Broad Institute Harvard MIT 272 454569 10 Bert Vogelstein Johns Hopkins University 270 410260 Brigham and Women's Hospital 11 267 363862 Eugene Braunwald Harvard Medical School Ecole Polytechnique Fédérale de 12 264 364838 Michael Graetzel Lausanne 13 Frank B Hu Harvard University 256 307111 14 Yi Hwa Liu Yale University 255 332019 15 M A Caligiuri City of Hope National Medical Center 253 345173 16 Gordon Guyatt McMaster University 252 284725 17 Salim Yusuf McMaster University 250 357419 18 Michael Karin University of California San Diego 250 273000 Yale University; Howard Hughes 19 244 221895 Richard A Flavell Medical Institute 20 T W Robbins University of Cambridge 239 180615 21 Zhong Lin Wang Georgia Institute of Technology 238 234085 22 Martín Heidegger Universität Freiburg 234 335652 23 Paul M Ridker Harvard Medical School 234 318801 24 Daniel Levy National Institutes of Health NIH 232 286694 25 Guido Kroemer INSERM 231 240372 26 Steven A Rosenberg National Institutes of Health NIH 231 224154 Max Planck
    [Show full text]
  • Some Directed Graph Algorithms and Their Application to Pointer Analysis
    University of London Imperial College of Science, Technology and Medicine Department of Computing Some directed graph algorithms and their application to pointer analysis David J. Pearce February 2005 Submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Engineering of the University of London Abstract This thesis is focused on improving execution time and precision of scalable pointer analysis. Such an analysis statically determines the targets of all pointer variables in a program. We formulate the analysis as a directed graph problem, where the solution can be obtained by a computation similar, in many ways, to transitive closure. As with transitive closure, identifying strongly connected components and transitive edges offers significant gains. However, our problem differs as the computation can result in new edges being added to the graph and, hence, dynamic algorithms are needed to efficiently identify these structures. Thus, pointer analysis has often been likened to the dynamic transitive closure problem. Two new algorithms for dynamically maintaining the topological order of a directed graph are presented. The first is a unit change algorithm, meaning the solution must be recomputed immediately following an edge insertion. While this has a marginally inferior worse-case time bound, compared with a previous solution, it is far simpler to implement and has fewer restrictions. For these reasons, we find it to be faster in practice and provide an experimental study over random graphs to support this. Our second is a batch algorithm, meaning the solution can be updated after several insertions, and it is the first truly dynamic solution to obtain an optimal time bound of O(v + e + b) over a batch b of edge insertions.
    [Show full text]
  • Diadem and Veriware Press Release
    - 1 - 2nd November 2009 Oxford Computer Scientists Win Prestigious ERC Awards Two professors at the Oxford University Computing Laboratory have been selected to receive highly competitive European grants for research projects that promise major breakthroughs in how we use technology. The European Research Council has made Advanced Investigators Grants worth a total of €4.4m for the Diadem project, headed by Georg Gottlob, and for Veriware under Marta Kwiatkowska. Fewer than ten of these grants, intended to support pioneering research by leaders in their fields, have been made this year to computer scientists across Europe. The projects will both last five years, and address challenging but vital research questions that will be central to our relationship with the technology around us over the coming decades. One will help ensure the electronic devices we rely on work properly; the other will help us extract the information we need from the ever-increasing volume of data on the internet by enabling computers to understand its structure on their own. With Veriware, Kwiatkowska aims to make the theoretical and practical breakthroughs that will let us be sure that our technology is functioning as it should as we enter a new era of ubiquitous computing, in which information processing moves out of desktop computers and into the everyday objects all around us. This kind of computing without computers has been called ‘everyware’. The technology is still in its infancy, but examples are starting to appear – from Bluetooth mobile phones that automatically sense each others’ presence and exchange information to fridges that can tell when the food in them is running out and automatically order more from the internet.
    [Show full text]
  • Safety Verification for Deep Neural Networks with Provable
    Safety Verification for Deep Neural Networks with Provable Guarantees Marta Z. Kwiatkowska Department of Computer Science, University of Oxford, UK http://www.cs.ox.ac.uk/marta.kwiatkowska/ [email protected] Abstract Computing systems are becoming ever more complex, increasingly often incorporating deep learning components. Since deep learning is unstable with respect to adversarial perturbations, there is a need for rigorous software development methodologies that encompass machine learning. This paper describes progress with developing automated verification techniques for deep neural networks to ensure safety and robustness of their decisions with respect to input perturbations. This includes novel algorithms based on feature-guided search, games, global optimisation and Bayesian methods. 2012 ACM Subject Classification Theory of computation → Logic and verification; Computing methodologies → Neural networks Keywords and phrases Neural networks, robustness, formal verification, Bayesian neural networks Digital Object Identifier 10.4230/LIPIcs.CONCUR.2019.1 Category Invited Paper Funding Marta Z. Kwiatkowska: This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 834115) and the EPSRC Programme Grant on Mobile Autonomy (EP/M019918/1). 1 Introduction Computing devices have become ubiquitous and ever present in our lives: smartphones help us stay in touch with family and friends, GPS-enabled apps offer directions literally at our fingertips, and voice-controlled assistants are now able to execute simple commands. Artificial Intelligence is making great strides, promising many more exciting applications with an increased level of autonomy, from wearable medical devices to robotic care assistants and self-driving cars.
    [Show full text]
  • Prototyping Parallel Functional Intermediate Languages
    Prototyping parallel functional intermediate languages by Andrew David Ben-Dyke A thesis submitted to the Faculty of Science of The University of Birmingham for the degree of DOCTOR OF PHILOSOPHY School of Computer Science Faculty of Science The University of Birmingham United Kingdom October 1999 University of Birmingham Research Archive e-theses repository This unpublished thesis/dissertation is copyright of the author and/or third parties. The intellectual property rights of the author or third parties in respect of this work are as defined by The Copyright Designs and Patents Act 1988 or as modified by any successor legislation. Any use made of information contained in this thesis/dissertation must be in accordance with that legislation and must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the permission of the copyright holder. Abstract Non-strict higher-order functional programming languages are elegant, concise, math­ ematically sound and contain few environment-specific features, making them obvious candidates for harnessing high-performance architectures. The validity of this approach has been established by a number of experimental compilers. However, while there have been a number of important theoretical developments in the field of parallel functional programming, implementations have been slow to materialise. The myriad design choices and demands of specific architectures lead to protracted development times. Furthermore, the resulting systems tend to be monolithic entities, and are difficult to extend and test, ultimatly discouraging experimentation. The traditional solution to this problem is the use of a rapid prototyping framework. However, as each existing systems tends to prefer one specific platform and a particular way of expressing parallelism (including implicit specification) it is difficult to envisage a general purpose framework.
    [Show full text]