DOCSLIB.ORG

Inspiredresearch

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text

Load more

InspiredResearch Summer 2016 Issue 8 News from the Department of Computer Science, University of Oxford TheThe futurefuture ofof ArtificialArtificial IntelligenceIntelligence Google DeepMind’s Demis Hassabis gives Strachey Lecture – p26 In search of the New Forest cicada: Developing new tools for acoustic ecology – p6 Preventing lunchtime attacks: Fighting insider threats with eye movement biometrics – p12 Robots and the moral maze: Towards a code of ethics for artificial intelligence – p23 CompSciOxford DepartmentOfComputerScienceUniversityOfOxford Inspired Research Letter from the is a twice-yearly newsletter published by the Department of Computer Science at the University Head of Department of Oxford. If you would like to learn more One of the many rewards of academic life at Oxford is the constant about anything you read in these stream of incredible academic speakers that pass through our doors. A pages, please get in touch: quick look at our forthcoming seminar series tells me that in the coming [email protected] week, I can hear about ‘The expressiveness of decidable fixpoint logics’, ‘Fighting modern cyber criminals’, and ‘Semantic alignment for agent To subscribe to future issues, interactions’. This is not a phenomenon restricted to the Department e-mail: [email protected] of Computer Science, of course: Oxford is one of the world’s great To download previous issues, visit intellectual meeting places, and a visit to Oxford is very like a pilgrimage www.cs.ox.ac.uk/inspiredresearch for most academics. Editorial board Of all the remarkable speakers I have had the pleasure to see at Oxford, Suzanna Marsh (Editor) none has been more remarkable than Demis Hassabis, who gave the Hilary Term Strachey Lecture in February 2016 (see page 26). Demis is a Kiri Walden (Sub-Editor) founder and CEO of DeepMind, the London-based artificial intelligence Suvarna Designs (Designer) company that was acquired by Google in 2014 for $400m. We invited Demis to give this lecture a year ago, and had a strong hunch that it Benjamin Neal was going to be a popular event. In anticipation of this, and thanks to Daniel Marsden generous financial support from Oxford Asset Management, we were Daniel Neville able to book Christopher Wren’s Sheldonian Theatre for the lecture, with Dima Pasechnik a reception afterwards in the Divinity School. The stage was set for a special occasion. However, nothing prepared us for the deluge of interest Emma Dunlop that Demis’s lecture generated. All 800 tickets were taken within 48 hours Faris Abou-Saleh of release, and we were so overwhelmed by pleas for additional tickets Helena Webb that we booked another 300-seat lecture theatre and had the event live Jason Nurse streamed there. Martha Lewis On the day, Demis arrived complete with a camera crew making a film Maureen York of his life, and I had the very odd experience of walking through Oxford Peter Jeavons with Demis making smalltalk, while a camera crew filmed us from across Sarah Baldwin the road. (A PhD in Computer Science doesn’t prepare you for things like Seyyed Shah this…) The lecture itself was a joy – a model of what a Strachey Lecture should be, pitched perfectly, with plenty of good humour and much Shoshannah Holdom food for thought. So, you are probably asking yourself, why all the fuss? Thomas Lukasiewicz Well, DeepMind’s core technology is deep learning – probably the most Will Smith active research area in Computer Science at present. Deep learning has made huge progress over the past decade, and no clearer evidence of this is provided by the fact that, just a week after Demis’s Oxford lecture, Contributors DeepMind triumphed in a series of matches in which a self-taught AI Alex Kavvos Monica Wilson programme comprehensively defeated one of the world’s best Go players. Alex Rogers Nathan Walk The series of matches made headlines across the globe. With Demis Andrew Martin Niel de Beaudrap giving his Oxford lecture just a week before the Go tournament, everyone present at the lecture felt they were witnessing history in the making. And Bushra AlAhmadi Paula Boddington indeed, they were. David Gavaghan Reuben Binns David Hobbs Sam Littley In other exciting news, I’m delighted to be able to Helena Webb Sharon Lloyd say that Sir Jonathan Ive, head of design at Apple, Ivan Martinovic Shimon Whiteson Inc., will receive an honorary degree at the annual Jassim Happa Stephen Pulman encaenia event in June (page 3). Sir Jonathan is Katherine Fletcher Troy Astarte probably the most influential industrial designer of Laura Jones Ursula Martin the past 50 years: amongst other things, he designed Lucy Crittenden the iPad and iPhone. He designed the iPod! Matty Hoban Michael Wooldridge Professor Michael Wooldridge May 2016 2 Inspired Research: News from the Department of Computer Science, University of Oxford NEWS Apple’s Chief Design for a range of Apple products, including the ground-breaking News in brief Officer Sir Jonathan original iMac in the late 1990s, the Ive to receive iPod, and the iPhone. Apart from being commercially successful on honorary degree a truly global scale, his trailblazing designs have dramatically influenced not just the look and feel of these We are delighted to announce that devices, but the way we think Sir Jonathan Ive, the Chief Design about and interact with technology Officer at Apple, Inc., will receive an itself. Perhaps the best way we can honorary degree from the University summarise Sir Jonathan’s influence of Oxford at the University’s annual would be to point out that most Encaenia ceremony, this year being people reading this announcement DPhil student Elizabeth Phillips has held on 22 June 2016. online will do so on a device either been selected as the only student designed or directly influenced by alongside nine UK government Sir Jonathan’s nomination from the him. officials to participate in a US Department of Computer Science government-sponsored exchange recognises his iconic, hugely We very much look forward to programme to identify the top influential, and incredibly successful welcoming Sir Jonathan to Oxford ten leaders in UK Cyber Security. contributions to industrial design. in June. Participants were selected by Sir Jonathan was the chief designer the US Department of State and the US Embassy in London. The programme enables participants to learn about US efforts to protect digital infrastructure, ensure free flow of information, security and privacy of data, and the integrity of networks. Elizabeth met President Obama at the ‘Town Hall’ he hosted in London in April. Oxford involved explore issues in privacy, ethics, of our daily lives will be connected, trust, reliability, acceptability, and in one way or another, to the digital in new Internet security (known as PETRAS). world. Physical objects and devices of Things (IoT) will be able to interact with each The project is part of IoTUK, an other, ourselves, and the wider Research Hub integrated £40 million initiative virtual world. But, before this can between government, industry, happen, there must be trust and the research community and the confidence in how the Internet Department of Computer Science public sector to help advance UK of Things works, its security and academics, Professors Andrew global leadership in the IoT and its resilience. By harnessing our Martin, Sadie Creese and Kasper increase adoption of high quality world-leading research excellence Rasmussen are involved in the IoT technologies and services by this PETRAS research hub newly announced £23 million businesses and the public sector in will accelerate IoT technology Internet of Things (IoT) Research the UK, benefiting citizens. The hub innovation and bring benefit[s] to Hub. They have been working with will draw in substantial support from society and business.’ fellow Oxford Professors, Luciano over 47 partners from industry and Floridi of the Oxford Internet the public sector. Institute and David De Roure of the Oxford e-Research Centre. Professor Philip Nelson, the Engineering Physical and Live Oxford is one of a consortium of Sciences Research Council’s nine UK universities which will (EPSRC) Chief Executive, said, ‘In work over a three-year period to the not too distant future almost all www.cs.ox.ac.uk Issue 8 Summer 2016 3 NEWS News in brief Professor Bill Roscoe wins MPLS Oxford climbed to top spot in The Sunday Times University Lifetime Award Guide 2016 league table for Computer Science. It also took Professor Bill Roscoe has been third place (best in the UK) in the given a lifetime award by the Oxford QS World University Rankings University Mathematical, Physical 2016 for Computer Science and and Life Sciences Division for Information Systems. In the Times external engagement and promoting Higher Education World University impact. Rankings Oxford has consolidated its position as one of the world’s The former Head of Department has top universities. Overall Oxford been recognised for his pioneering was the top European University work applying formal verification in the rankings (2nd in the world), tools in industry since the late 1980s, and in subject-specific tables for including the invention of the FDR Engineering and Technology the Software Verification Tool. University was rated as 2nd in theoretical research. Furthermore, Europe and 6th in the world. He says of his award, ‘It has been interaction with these users has been wonderful to see so many real-world a regular source of inspiration for my The Alan Turing Institute (ATI) held users of what originally seemed like research: a real virtuous circle!’ a symposium on reproducibility for data-intensive research. Reproducibility is both a technical Degree application numbers climb and a socio-cultural issue, requiring new metadata systems, technical Application figures for the Oxford undergraduate Computer Science degree architectures, workflows and programmes have risen for the 9th consecutive year.
Recommended publications
  • Lecture Note on Deep Learning and Quantum Many-Body Computation

    Lecture Note on Deep Learning and Quantum Many-Body Computation

    Lecture Note on Deep Learning and Quantum Many-Body Computation Jin-Guo Liu, Shuo-Hui Li, and Lei Wang∗ Institute of Physics, Chinese Academy of Sciences Beijing 100190, China November 23, 2018 Abstract This note introduces deep learning from a computa- tional quantum physicist’s perspective. The focus is on deep learning’s impacts to quantum many-body compu- tation, and vice versa. The latest version of the note is at http://wangleiphy.github.io/. Please send comments, suggestions and corrections to the email address in below. ∗ [email protected] CONTENTS 1 introduction2 2 discriminative learning4 2.1 Data Representation 4 2.2 Model: Artificial Neural Networks 6 2.3 Cost Function 9 2.4 Optimization 11 2.4.1 Back Propagation 11 2.4.2 Gradient Descend 13 2.5 Understanding, Visualization and Applications Beyond Classification 15 3 generative modeling 17 3.1 Unsupervised Probabilistic Modeling 17 3.2 Generative Model Zoo 18 3.2.1 Boltzmann Machines 19 3.2.2 Autoregressive Models 22 3.2.3 Normalizing Flow 23 3.2.4 Variational Autoencoders 25 3.2.5 Tensor Networks 28 3.2.6 Generative Adversarial Networks 29 3.3 Summary 32 4 applications to quantum many-body physics and more 33 4.1 Material and Chemistry Discoveries 33 4.2 Density Functional Theory 34 4.3 “Phase” Recognition 34 4.4 Variational Ansatz 34 4.5 Renormalization Group 35 4.6 Monte Carlo Update Proposals 36 4.7 Tensor Networks 37 4.8 Quantum Machine Leanring 38 4.9 Miscellaneous 38 5 hands on session 39 5.1 Computation Graph and Back Propagation 39 5.2 Deep Learning Libraries 41 5.3 Generative Modeling using Normalizing Flows 42 5.4 Restricted Boltzmann Machine for Image Restoration 43 5.5 Neural Network as a Quantum Wave Function Ansatz 43 6 challenges ahead 45 7 resources 46 BIBLIOGRAPHY 47 1 1 INTRODUCTION Deep Learning (DL) ⊂ Machine Learning (ML) ⊂ Artificial Intelli- gence (AI).
  • AI Computer Wraps up 4-1 Victory Against Human Champion Nature Reports from Alphago's Victory in Seoul

    AI Computer Wraps up 4-1 Victory Against Human Champion Nature Reports from Alphago's Victory in Seoul

    The Go Files: AI computer wraps up 4-1 victory against human champion Nature reports from AlphaGo's victory in Seoul. Tanguy Chouard 15 March 2016 SEOUL, SOUTH KOREA Google DeepMind Lee Sedol, who has lost 4-1 to AlphaGo. Tanguy Chouard, an editor with Nature, saw Google-DeepMind’s AI system AlphaGo defeat a human professional for the first time last year at the ancient board game Go. This week, he is watching top professional Lee Sedol take on AlphaGo, in Seoul, for a $1 million prize. It’s all over at the Four Seasons Hotel in Seoul, where this morning AlphaGo wrapped up a 4-1 victory over Lee Sedol — incidentally, earning itself and its creators an honorary '9-dan professional' degree from the Korean Baduk Association. After winning the first three games, Google-DeepMind's computer looked impregnable. But the last two games may have revealed some weaknesses in its makeup. Game four totally changed the Go world’s view on AlphaGo’s dominance because it made it clear that the computer can 'bug' — or at least play very poor moves when on the losing side. It was obvious that Lee felt under much less pressure than in game three. And he adopted a different style, one based on taking large amounts of territory early on rather than immediately going for ‘street fighting’ such as making threats to capture stones. This style – called ‘amashi’ – seems to have paid off, because on move 78, Lee produced a play that somehow slipped under AlphaGo’s radar. David Silver, a scientist at DeepMind who's been leading the development of AlphaGo, said the program estimated its probability as 1 in 10,000.
  • Accelerators and Obstacles to Emerging ML-Enabled Research Fields

    Accelerators and Obstacles to Emerging ML-Enabled Research Fields

    Life at the edge: accelerators and obstacles to emerging ML-enabled research fields Soukayna Mouatadid Steve Easterbrook Department of Computer Science Department of Computer Science University of Toronto University of Toronto [email protected] [email protected] Abstract Machine learning is transforming several scientific disciplines, and has resulted in the emergence of new interdisciplinary data-driven research fields. Surveys of such emerging fields often highlight how machine learning can accelerate scientific discovery. However, a less discussed question is how connections between the parent fields develop and how the emerging interdisciplinary research evolves. This study attempts to answer this question by exploring the interactions between ma- chine learning research and traditional scientific disciplines. We examine different examples of such emerging fields, to identify the obstacles and accelerators that influence interdisciplinary collaborations between the parent fields. 1 Introduction In recent decades, several scientific research fields have experienced a deluge of data, which is impacting the way science is conducted. Fields such as neuroscience, psychology or social sciences are being reshaped by the advances in machine learning (ML) and the data processing technologies it provides. In some cases, new research fields have emerged at the intersection of machine learning and more traditional research disciplines. This is the case for example, for bioinformatics, born from collaborations between biologists and computer scientists in the mid-80s [1], which is now considered a well-established field with an active community, specialized conferences, university departments and research groups. How do these transformations take place? Do they follow similar paths? If yes, how can the advances in such interdisciplinary fields be accelerated? Researchers in the philosophy of science have long been interested in these questions.
  • The Deep Learning Revolution and Its Implications for Computer Architecture and Chip Design

    The Deep Learning Revolution and Its Implications for Computer Architecture and Chip Design

    The Deep Learning Revolution and Its Implications for Computer Architecture and Chip Design Jeffrey Dean Google Research [email protected] Abstract The past decade has seen a remarkable series of advances in machine learning, and in particular deep learning approaches based on artificial neural networks, to improve our abilities to build more accurate systems across a broad range of areas, including computer vision, speech recognition, language translation, and natural language understanding tasks. This paper is a companion paper to a keynote talk at the 2020 International Solid-State Circuits Conference (ISSCC) discussing some of the advances in machine learning, and their implications on the kinds of computational devices we need to build, especially in the post-Moore’s Law-era. It also discusses some of the ways that machine learning may also be able to help with some aspects of the circuit design process. Finally, it provides a sketch of at least one interesting direction towards much larger-scale multi-task models that are sparsely activated and employ much more dynamic, example- and task-based routing than the machine learning models of today. Introduction The past decade has seen a remarkable series of advances in machine learning (ML), and in particular deep learning approaches based on artificial neural networks, to improve our abilities to build more accurate systems across a broad range of areas [LeCun et al. 2015]. Major areas of significant advances ​ ​ include computer vision [Krizhevsky et al. 2012, Szegedy et al. 2015, He et al. 2016, Real et al. 2017, Tan ​ ​ ​ ​ ​ ​ and Le 2019], speech recognition [Hinton et al.
  • The Power and Promise of Computers That Learn by Example

    The Power and Promise of Computers That Learn by Example

    Machine learning: the power and promise of computers that learn by example MACHINE LEARNING: THE POWER AND PROMISE OF COMPUTERS THAT LEARN BY EXAMPLE 1 Machine learning: the power and promise of computers that learn by example Issued: April 2017 DES4702 ISBN: 978-1-78252-259-1 The text of this work is licensed under the terms of the Creative Commons Attribution License which permits unrestricted use, provided the original author and source are credited. The license is available at: creativecommons.org/licenses/by/4.0 Images are not covered by this license. This report can be viewed online at royalsociety.org/machine-learning Cover image © shulz. 2 MACHINE LEARNING: THE POWER AND PROMISE OF COMPUTERS THAT LEARN BY EXAMPLE Contents Executive summary 5 Recommendations 8 Chapter one – Machine learning 15 1.1 Systems that learn from data 16 1.2 The Royal Society’s machine learning project 18 1.3 What is machine learning? 19 1.4 Machine learning in daily life 21 1.5 Machine learning, statistics, data science, robotics, and AI 24 1.6 Origins and evolution of machine learning 25 1.7 Canonical problems in machine learning 29 Chapter two – Emerging applications of machine learning 33 2.1 Potential near-term applications in the public and private sectors 34 2.2 Machine learning in research 41 2.3 Increasing the UK’s absorptive capacity for machine learning 45 Chapter three – Extracting value from data 47 3.1 Machine learning helps extract value from ‘big data’ 48 3.2 Creating a data environment to support machine learning 49 3.3 Extending the lifecycle
  • ARCHITECTS of INTELLIGENCE for Xiaoxiao, Elaine, Colin, and Tristan ARCHITECTS of INTELLIGENCE

    ARCHITECTS of INTELLIGENCE for Xiaoxiao, Elaine, Colin, and Tristan ARCHITECTS of INTELLIGENCE

    MARTIN FORD ARCHITECTS OF INTELLIGENCE For Xiaoxiao, Elaine, Colin, and Tristan ARCHITECTS OF INTELLIGENCE THE TRUTH ABOUT AI FROM THE PEOPLE BUILDING IT MARTIN FORD ARCHITECTS OF INTELLIGENCE Copyright © 2018 Packt Publishing All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews. Every effort has been made in the preparation of this book to ensure the accuracy of the information presented. However, the information contained in this book is sold without warranty, either express or implied. Neither the author, nor Packt Publishing or its dealers and distributors, will be held liable for any damages caused or alleged to have been caused directly or indirectly by this book. Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy of this information. Acquisition Editors: Ben Renow-Clarke Project Editor: Radhika Atitkar Content Development Editor: Alex Sorrentino Proofreader: Safis Editing Presentation Designer: Sandip Tadge Cover Designer: Clare Bowyer Production Editor: Amit Ramadas Marketing Manager: Rajveer Samra Editorial Director: Dominic Shakeshaft First published: November 2018 Production reference: 2201118 Published by Packt Publishing Ltd. Livery Place 35 Livery Street Birmingham B3 2PB, UK ISBN 978-1-78913-151-2 www.packt.com Contents Introduction ........................................................................ 1 A Brief Introduction to the Vocabulary of Artificial Intelligence .......10 How AI Systems Learn ........................................................11 Yoshua Bengio .....................................................................17 Stuart J.
  • Technology Innovation and the Future of Air Force Intelligence Analysis

    Technology Innovation and the Future of Air Force Intelligence Analysis

    C O R P O R A T I O N LANCE MENTHE, DAHLIA ANNE GOLDFELD, ABBIE TINGSTAD, SHERRILL LINGEL, EDWARD GEIST, DONALD BRUNK, AMANDA WICKER, SARAH SOLIMAN, BALYS GINTAUTAS, ANNE STICKELLS, AMADO CORDOVA Technology Innovation and the Future of Air Force Intelligence Analysis Volume 2, Technical Analysis and Supporting Material RR-A341-2_cover.indd All Pages 2/8/21 12:20 PM For more information on this publication, visit www.rand.org/t/RRA341-2 Library of Congress Cataloging-in-Publication Data is available for this publication. ISBN: 978-1-9774-0633-0 Published by the RAND Corporation, Santa Monica, Calif. © Copyright 2021 RAND Corporation R® is a registered trademark. Cover: U.S. Marine Corps photo by Cpl. William Chockey; faraktinov, Adobe Stock. Limited Print and Electronic Distribution Rights This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited. Permission is given to duplicate this document for personal use only, as long as it is unaltered and complete. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial use. For information on reprint and linking permissions, please visit www.rand.org/pubs/permissions. The RAND Corporation is a research organization that develops solutions to public policy challenges to help make communities throughout the world safer and more secure, healthier and more prosperous. RAND is nonprofit, nonpartisan, and committed to the public interest. RAND’s publications do not necessarily reflect the opinions of its research clients and sponsors.
  • Machine Learning Conference Report

    Machine Learning Conference Report

    Part of the conference series Breakthrough science and technologies Transforming our future Machine learning Conference report Machine learning report – Conference report 1 Introduction On 22 May 2015, the Royal Society hosted a unique, high level conference on the subject of machine learning. The conference brought together scientists, technologists and experts from across academia, industry and government to learn about the state-of-the-art in machine learning from world and UK experts. The presentations covered four main areas of machine learning: cutting-edge developments in software, the underpinning hardware, current applications and future social and economic impacts. This conference is the first in a new series organised This report is not a verbatim record, but summarises by the Royal Society, entitled Breakthrough Science the discussions that took place during the day and the and Technologies: Transforming our Future, which will key points raised. Comments and recommendations address the major scientific and technical challenges reflect the views and opinions of the speakers and of the next decade. Each conference will focus on one not necessarily that of the Royal Society. Full versions technology and cover key issues including the current of the presentations can be found on our website at: state of the UK industry sector, future direction of royalsociety.org/events/2015/05/breakthrough-science- research and the wider social and economic implications. technologies-machine-learning The conference series is being organised through the Royal Society’s Science and Industry programme, which demonstrates our commitment to reintegrate science and industry at the Society and to promote science and its value, build relationships and foster translation.
  • Arxiv:1709.02779V1 [Quant-Ph] 8 Sep 2017 I

    Arxiv:1709.02779V1 [Quant-Ph] 8 Sep 2017 I

    Machine learning & artificial intelligence in the quantum domain Vedran Dunjko Institute for Theoretical Physics, University of Innsbruck, Innsbruck 6020, Austria Max Planck Institute of Quantum Optics, Garching 85748, Germany Email: [email protected] Hans J. Briegel Institute for Theoretical Physics, University of Innsbruck Innsbruck 6020, Austria Department of Philosophy, University of Konstanz, Konstanz 78457, Germany Email: [email protected] Abstract. Quantum information technologies, on the one side, and intelligent learning systems, on the other, are both emergent technologies that will likely have a transforming impact on our society in the future. The respective underlying fields of basic research { quantum information (QI) versus machine learning and artificial intelligence (AI) { have their own specific questions and challenges, which have hitherto been investigated largely independently. However, in a growing body of recent work, researchers have been prob- ing the question to what extent these fields can indeed learn and benefit from each other. QML explores the interaction between quantum computing and machine learning, inves- tigating how results and techniques from one field can be used to solve the problems of the other. In recent time, we have witnessed significant breakthroughs in both directions of influence. For instance, quantum computing is finding a vital application in providing speed-ups for machine learning problems, critical in our \big data" world. Conversely, machine learning already permeates many cutting-edge technologies, and may become instrumental in advanced quantum technologies. Aside from quantum speed-up in data analysis, or classical machine learning optimization used in quantum experiments, quan- tum enhancements have also been (theoretically) demonstrated for interactive learning tasks, highlighting the potential of quantum-enhanced learning agents.
  • Updated Compressed Data Trends

    Updated Compressed Data Trends

    Trends in Data Science 2019/2020 Foreword Across six major conferences in Boston, NYC, Sao Paulo, Bangalore, London, and San Francisco, ODSC hosted an unprecedented 920+ speakers in 2019. It was another year of rapid advances and exciting developments in the fields of data science and artificial intelli- gence. The year was significant on several fronts. Frameworks like PyTorch, Keras, and TensorFlow saw major releases that further cemented their position as the leading machine and deep learning tools and featured in many of our hands-on training sessions. However, NLP transformer architectures was one of our speak- ers’ favorite topics in 2019 due to major advances such as OpenAI’s GPT-2, Google’s BERT, DiDI’s Elmo, and FaceBook’s RoBERTa. Hugging Face’s transformer library which exposes an API to these transformers got a lot of traction also. These pretrained models, especially BERT, have been especially hailed by many as NLPs Imagenet moment. Employing new techniques like bi-directional sequencing and transformers these models are saving data scientists the time and expenses normally required to train NLP models thus marking a major development. Our new MLOps and Data Engineering focus area tracks coincided with the massive ramp up of efforts to increase the percent of data science projects into production. Tools like MLFlow and Kubeflow continued to grow in popularity. Interest in AutoML grew exponen- tially in 2019, given its potential as a productivity tool in all stages of the machine learning life cycle. Sessions around Annotation, model interoperability, pipelines, deployment, and testing saw increased interest. This coupled with the fact that 2019 saw a significant drop in the cost of modeling helped accelerate deployment in production environments.
  • Oxford (UK), the Entrance to Broad Street, at Catte Street (East) End

    Oxford (UK), the Entrance to Broad Street, at Catte Street (East) End

    Oxford (UK), the entrance to Broad Street, at Catte Street (east) end, looking towards Radcliffe Square (©Ozeye 2009) Oxford (UK), view of the East end of Broad Street, with the Weston Library on the left, Oxford Martin School (Indian Institute) straight ahead, and the Bodleian Libraries' Clarendon Building on the right (© Bodleian Libraries Communications team (John Cairns) 2015) Oxford (UK), Broad Street, showing the Clarendon Building, Sheldonian Theatre and the History of Science Museum (©Komarov 2016) Oxford (UK), Broad Street, looking towards the Clarendon Building and Sheldonian Theatre (©Linsdell 1989) Oxford (UK), Broad Street, looking towards the Clarendon Building (©Elbers 2018) Oxford (UK), Broad Street, looking towards the Sheldonian Theatre, Clarendon Building in the background (©Addison 2010) Oxford (UK), Broad Street, Sheldonian Theatre (©Ozeye 2009) Oxford (UK), Broad Street, the History of Science Museum (©Hawgood 2009) Oxford (UK), Broad Street, the History of Science Museum (©Wiki alf 2006) Oxford (UK), Broad Street, Blackwells as viewed from the History of Science Museum opposite on Broad Street (©Ozeye 2009) Oxford (UK), Broad Street, looking towards the Sheldonian Theatre and the History of Science Museum (©Trimming 2009) Oxford (UK), Broad Street, floor plan of the area around the Clarendon building, Sheldonian Theatre, the History of Science Museum and the opposite the Bodleian Weston Library where the ‘History and Mystery’ of the Sheldonian Heads exhibition will take place in 2019 (©HistoricEngland 2018) Plan of Sheldonian and some of the Bodleian complex showing quadrangles created by the placement of the Sheldonian(©University of Oxford 2012, New Bodleian Library: Design and Access Statement, Wilkinson Eyre Associates (March, 2010)) Heads at the Hartcourt Arboretum, Oxford University‘s botanic garden (potentially 2nd generation from the History of Science Museum.
  • The Building of the Second Palace at Cuddesdon

    The Building of the Second Palace at Cuddesdon

    The Building of the Second Palace at Cuddesdon By J. C. COLE N this paper I propose to discuss the surviving documents connected with I two legal disputes which arose in Oxford during the second half of the seventeenth century, from which we can learn some details of the building of the second palace at Cuddesdon and the craftsmen who were employed upon that work. The first of these disputes was brought before the Court of Arches in ,66g,' the second before the Vice-Chancellor's Court in ,681.' Before 1634 the Bishops of Oxford had no dwelling house especially appropriated to their use, but lived either in their parsonage houses or in hired lodgings in Oxford. In that year, as Anthony Wood tells us, William Laud, then Archbishop, persuaded the Bishop of Oxford, John Bancroft, to build a house for his own use and that of his successors' for ever '.3 The site chosen was the small village of Cuddesdon, of which Bishop Bancroft happened to hold the incumbency. The place was conveniently situated about five miles to the south-east of Oxford and not far from the old London Road. The building, which displaced an earlier parsonage house described as mean and ruinous,' was said to have cost about £2,600. King Charles gave his approval to the project and contributed fifty timber trees from the royal forest of Shot­ over as well as remitting a sum of £343 from the first fruits of the bishopric.' Several representations are to be found of the palace, which contemporaries called' a fair house of stone '.6 Laud paid it a visit of inspection in ,635' and stayed there again in ,636 on his way from London to Oxford to entertain the King.