Loading…
PAPIs Europe 2018 has ended
PAPIs is the 1st series of international conferences dedicated to real-world ML applications, techniques and tools. After 7 previous events on 4 different continents, PAPIs is returning to Europe on 4-6 April 2018. Join us at the Canary Wharf Tower in London!

  • Training Workshops — 4 April, UCL School of Management (Level 38)
  • Industry and Startups — 5 April, Level 39
  • Industry and Research — 6 April, Level 39

More information about the conference at papis.io/europe-2018

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Wednesday, April 4
 

09:00 BST

"GDPR: is your ML system ready?" — Workshop brought to you by Ercom
This training workshop will take place before the main conference. It will be given in a small classroom, to maximize interaction and so you can ask even more questions than in a conference setting. This is an agnostic workshop which is supported by Ercom; they develop simple and secured business solutions for enterprise file sharing and mobile communications.
-----

Implementing machine learning applications with interpretability, accountability, and trust

The EU General Data Protection Regulation (GDPR) enters into effect in May 2018. Its enforcement will dramatically change the way European companies handle data relating to individuals, and global companies handle data relating to EU residents. This is a big step towards more transparency over data, however, it implies several major challenges for data science practitioners. In particular it brings a need to understand how machine learning systems are profiling users, the avoidance of bias and discrimination on the basis of categories of data such as racial origin or political opinion. Particular focus has been given to the inclusion of a “right to an explanation” which sits alongside older requirements to inform users about the logics of automated processing which significantly affects them. Ensuring that machine learning systems are sufficiently transparent, and that they are not inferring sensitive characteristics without legal ground to do so, will impact the entire machine learning pipeline, and especially high dimensional machine learning practice and systems utilising inherently opaque methods such as deep neural networks and model ensembles (e.g. XGBoost).

Start preparing now your data science skills required to the be compliant with GDPR so that you can ensure your business doesn’t get behind.
You will learn:
  • how the GDPR rules will affect the way data teams do their work;
  • how the regulations will evolve over time, and how to best prepare for that now;
  • state-of-the-art safeguards against machine learning bias and discrimination;
  • how to use new machine learning procedures that ensure predictive power, explainability and trust to automated decisions, such as the creation of interpretable models, and how to apply techniques like LIME to audit and explain black-box models. 
Moreover, you will gain straightforward insights to demonstrate the behavior of a predictive model to stakeholders and regulators which brings fair and transparent decisions.


Outline
First part: GDPR overview
  • How to address immediate challenges for data teams: data governance, data protection, privacy by design, data subject requests (right to be forgotten, access, data portability, justification of decisions decided by an algorithm)
  • The impact of GDPR on algorithms
  • How to close the gaps to become GDPR compliant from data science perspective

Second part : Applied interpretable and fairness-aware machine learning, includes hands-on labs
  • What is interpretability and why it is important
  • De-biasing, the state-of-the-art and its limitations
  • Interpretable models
  • Explaining black-box models


Intended audience 
  • You are a Data Scientist, Data Analyst, aspiring or experienced, who wants to learn implications of GDPR in your data science process and how to tackle them.
  • You are a Data Engineer or Developer with some Machine Learning exposure who wants to learn how to implement typical compliant Data Science and ML workflows.
  • You are a Data Science or Data Engineering Manager, CTO, CIO or technical decision maker who aims to better understand the impact of GDPR on the data-driven enterprise.


Prerequisites
Some experience coding in Python or R and a basic understanding of data science topics and terminology are recommended. Experience using with data processing, feature generation, statistical modeling, and most common machine learning algorithms (linear regressions, trees) is helpful.


Hardware and/or installation requirements
The course will be run in R. Attendees should being a laptop with a recent R installation. An IDE such as RStudio would be useful. Attendees who wish to use Python may do so, and some resources will be provided for this. The first part of the course has no computing requirements.


Speakers
avatar for Michael Veale

Michael Veale

Researcher, UCL Department of Science, Technology, Engineering & Public Policy
I’m a technology policy researcher at University College London. I research:on-the-ground issues and design challenges for fairness, transparency and resilience of high-stakes algorithmic systems, particularly in the public sector;machine learning and privacy-enhancing technologies... Read More →


Wednesday April 4, 2018 09:00 - 17:30 BST
Level38 - UCL Seminar Suite Level38, One Canada Square, London

09:00 BST

Machine Learning Kickstart Workshop
This is a training workshop taking place before the main conference. It will be given in a classroom of up to 20 persons only, to maximize interaction and so you can ask even more questions than in a conference setting.
-----

Gain the skills to create Machine Learning models and use them in your applications, with open source libraries and cloud platforms. In this workshop, we'll demystify Machine Learning, you'll gain an understanding of its possibilities and limitations, and how to put it to work on real problems. You'll learn to create, evaluate and deploy ML models. We'll adopt a top-down, results-first and experimentation-driven approach.

Learning objectives
  • Understand the possibilities and limitations of ML
  • Understand the main ideas behind the most widely used learning algorithms in the industry: decision trees and random forests
  • Know how to build predictive models from data, analyze their performance and deploy to production with APIs.

Program

Each section contains theory and hands-on demos/exercises (where students can reproduce at the same time on their own laptops).

Introduction to Machine Learning
  • Key ML concepts and terminology
  • Formalizing supervised learning problems: classification and regression
  • Possibilities and example use cases (web applications, mobile, enterprise data science)

Model creation
  • Intuitions behind learning algorithms: Nearest Neighbors and Decision Trees
  • [Hands-on] Introduction to Jupyter notebooks
  • [Hands-on] Creating and interpreting Decision Trees with scikit-learn (open source ML library) and BigML (ML-as-a-Service tool), on classification and regression datasets

Evaluation
  • Performance criteria for ML models and evaluation procedure
  • Aggregate metrics for regression (MAE, MSE, R-squared, MAPE) and classification (accuracy, confusion and cost matrices)
  • [Hands-on] Evaluating models with Python, scikit-learn and BigML on previous datasets
  • [Hands-on] Improving prediction accuracy with ensembles of Decision Trees: Random Forests
  • [Hands-on] Embracing randomness with cross-validation

Deployment
  • [Hands-on] Why and how to use REST APIs for ML use in production
  • [Hands-on] Deploying your own Python models as APIs with the Flask library
  • [Hands-on] Using your API to fill in missing values in a spreadsheet program

Conclusions
  • Recap of key take-aways
  • Resources to go further

Student requirements

Programming experience and basic knowledge of the Python syntax. Code will be provided for students to replicate what will be shown during hands-on demos. Please consult Codeacademy's Learn Python and Robert Johansson's Introduction to Python programming (in particular the following sections: Python program files, Modules, Assignment, Fundamental types, Control Flow et Functions) to learn or revise Python's basics.
  • Basic maths knowledge (undergraduate level) will be useful to better understand some of the theory behind learning algorithms, but it isn’t a hard requirement.
  • Own laptop to bring for hands-on practical work.


Speakers
avatar for Sébastien Treguer

Sébastien Treguer

Lead Data Science, Tesobe / La Paillasse
Sebastien graduated from a master's degree in science with a specialization in signal and image processing. He has developed his experience in various fields, including finance, energy, healthcare, biology and neuroscience.His background in signal and image processing gave him the... Read More →



Wednesday April 4, 2018 09:00 - 17:30 BST
Level38 NE Theatre Level38, One Canada Square, London
 
Thursday, April 5
 

08:15 BST

Welcome & Registration
Thursday April 5, 2018 08:15 - 09:00 BST
Level39

09:00 BST

Opening Remarks
Alastair Moore (UCL), Florian Douetteau (Dataiku), Louis Dorard (PAPIs)

Thursday April 5, 2018 09:00 - 09:05 BST
Level39 - The Sandbox Level39, One Canada Square, London

09:05 BST

DIY Predictive Modeling Cluster with Kubernetes, Dask and JupyterHub.
In this presentation, I will show how to use Kubernetes to efficiently share computational resources (in the public cloud or on-premises) for a team of Data Scientists using Dask, Scikit-learn and JupyterHub.

Speakers
avatar for Olivier Grisel

Olivier Grisel

Engineer, INRIA / Scikit-learn
Olivier Grisel is a regular contributor to the scikit-learn library, he developed areas of expertise that include Machine Learning, Text Mining and Natural Language Processing. Olivier is currently employed as a Software Engineer at at Inria Parietal to work on improving scikit-learn... Read More →


Thursday April 5, 2018 09:05 - 09:45 BST
Level39 - The Sandbox Level39, One Canada Square, London

09:45 BST

Deep Learning at the edge with AWS DeepLens
AWS DeepLens is a wireless video camera and API that lets you get hands-on experience with Deep Learning and develop your own computer vision applications. In this talk, we'll look under the hood of Deep Lens and discuss its different building blocks: Intel hardware, Intel Inference Engine, Apache MXNet, AWS Greengrass and Amazon SageMaker. Demo included, of course!

Speakers
avatar for Julien Simon

Julien Simon

AI Evangelist, AWS
As the Artificial Intelligence & Machine Learning Evangelist for EMEA, Julien focuses on helping developers and enterprises bring their ideas to life. He's also actively blogging at https://medium.com/@julsimon. Prior to joining AWS, Julien served for 10 years as CTO/VP Engineering... Read More →


Thursday April 5, 2018 09:45 - 10:15 BST
Level39 - The Sandbox Level39, One Canada Square, London

10:15 BST

AI Startup Pitches
Thursday April 5, 2018 10:15 - 10:30 BST
Level39 - The Sandbox Level39, One Canada Square, London

10:30 BST

Expo - Networking - Tea/Coffee
Thursday April 5, 2018 10:30 - 11:00 BST
Level38 Level38, One Canada Square, London

11:00 BST

An Infinite Parade of Giraffes: AI Collaborative Cartooning
What is recognizable about a particular artist’s style? What parts can be delegated to an assistant? Can AI play the role of assistant? Of collaborator? How would we get enough training data? How little could we get away with?



Exploring these questions using GANs, image to image translation and extreme augmentation of very small data sets, a series of experiments in human/AI collaborative cartoon drawing provides a gentle introduction to machine learning and computer vision while presenting augmentation and rule based techniques for the more experienced user.

Speakers
avatar for Gretchen Greene

Gretchen Greene

Computer vision scientist/AI policy advisor, Greene Analytics/MIT Media Lab
Gretchen Greene, founder and CEO of Greene Strategy and Analytics, is a computer vision scientist, machine learning engineer and lawyer advising governments and private companies on AI use, strategy and policy. Greene has been interviewed by Forbes China, the Economist and the BBC... Read More →


Thursday April 5, 2018 11:00 - 11:30 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

11:00 BST

Changing Tires While Driving: Upgrading Algorithms Live!
Our industry is rapidly changing and we frequently need to upgrade our product to meet customer needs. This is non-trivial as our system runs 24x7 at huge scale, and uses unattended, automated machine learning to spend serious amounts of money every second.



In this presentation we describe a recent algorithm improvement to exploit a previously unanticipated situation. Although we will describe the algorithm, the presentation is focused on describing the pattern we have developed and perfected to safely upgrade our system and evaluate new algorithms while it is running.

Speakers
avatar for Beth Logan PhD

Beth Logan PhD

VP Optimization, dataxu
Beth is the VP of Optimization at DataXu, a leader in programmatic marketing. She has made contributions to a wide variety of fields, including speech recognition, music indexing and in-home activity monitoring. Beth holds a PhD in speech recognition from the University of Cambri... Read More →


Thursday April 5, 2018 11:00 - 11:30 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

11:00 BST

[Tutorial] From Zero to ML on Google Cloud
Zero to ML

Everyone’s talking about machine learning, but we hear much less about how to put it into practice. And let’s face it, that can be daunting! Even just 10 years ago you needed access to extensive academic and computing resources to make use of machine learning. Fast forward to today and we’ve seen revolutionary changes in the hardware and software that are making ML accessible for any developer or data scientist. Whether you’re completely new to ML or you’ve already trained and deployed your own model from scratch, Google Cloud Platform has a variety of tools to help you start using ML right now. I’ll start with the basics: how to use a pre-trained MLmodel with one REST API call. Then you’ll learn how to use your own dataset to customize a pre-trained model with transfer learning. We’ll end by learning how to build your own model from scratch with TensorFlow, and how train and serve it in the cloud with GCP.

Speakers
avatar for Daniel Bergqvist

Daniel Bergqvist

Developer Advocate, Google Cloud
Daniel has more than 5 years of experience as Cloud Platform Solutions Engineer at Google. Prior to working at Google, he was Head of Developer Relations at Blaast, a Finnish startup developing a cloud-powered mobile platform. With more than ten years of experience in the software... Read More →


Thursday April 5, 2018 11:00 - 12:00 BST
Level38 NE Theatre Level38, One Canada Square, London

11:30 BST

Music in the age of artificial creation
Machine learning has been making headlines with its sometimes alarming progress in skills previously thought to be the preserve of the human. What will become of music now that machines are capable of composing it? In our engineering and creative research, we are applying deep learning methods to model transcriptions of traditional music of Ireland and the UK. These models, trained on over 23,000 tunes, can then become creative partners in making new music. Several composers have worked with our models to compose dozens of new pieces, both within and outside the conventions of the traditional music on which the models were trained. This shows the exciting potential for such machines for augmenting human creativity, and opening new avenues for music practice.

Speakers
avatar for Oded Ben-Tal

Oded Ben-Tal

Senior Lecturer, Kingston University
Oded Ben-Tal is Senior Lecturer in Music Composition at the Department of Performing Arts, Kingston University. Ben-Tal specialises in music composition. Oded is a composer with complementary research interests at the intersection of Music, Cognition, and Computing. My compositions... Read More →
avatar for Bob Sturm

Bob Sturm

Lecturer, Queen Mary University of London
Sturm is a Lecturer in Digital Media at the School of Electronic Engineering and Computer Science, Queen Mary University of London. Sturm specialises in audio and music signal processing, machine listening, and evaluation.


Thursday April 5, 2018 11:30 - 12:00 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

11:30 BST

Scalable Machine Learning Pipelines for Click Predictions
Machine Learning is the cornerstone of ad-tech industry organizations. Building scalable pipelines for real-time bidding is crucial for buyers to compete for the most performing inventory on the Internet. We describe the Machine Learning pipeline that we built at AppNexus that operates globally and at Internet scales. The presented solutions describe a qualitative evaluation of our design choices, related to scalable processing infrastructure, high frequency model updates, and the pitfalls that must be avoided when interacting with low-latency model serving infrastructures for RTB.

Speakers
avatar for Moussa Taifi

Moussa Taifi

Data Science Platform Engineer, Appnexus
Moussa Taifi is currently a Senior Data Science Platform Engineer at AppNexus. He holds a PhD in Computer and Information Science from Temple university. His interests lie in large scale HPC and data-intensive parallel software systems, big data systems and applications, cloud technologies... Read More →


Thursday April 5, 2018 11:30 - 12:00 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

12:00 BST

Expo - Networking - Lunch
Thursday April 5, 2018 12:00 - 13:30 BST
Level38 Level38, One Canada Square, London

13:00 BST

[Workshop] How Can a Data Team Implement a GDPR-Compliant Project?
Use Case in the context of an e-commerce company implementing a recommender system
This talk details how data team leaders, data architects, and data scientists can organize themselves to comply with the GDPR for their data projects. In particular, we'll look at bundling a recommender system using two distinct situations: when data is anonymized, and when it is not.

Speakers
avatar for Alexandre Hubert

Alexandre Hubert

Lead Data Scientist UK, Dataiku
As data scientist, Alexandre has worked on a range of use cases, from creating models that predict fraud to building specific recommendation systems. He especially loves using deep learning with text or sports data. Even when he’s playing sport or having fun with friends, Alexandre... Read More →


Thursday April 5, 2018 13:00 - 14:30 BST
Level38 NE Theatre Level38, One Canada Square, London

13:30 BST

Solving the organizational challenge of predictive APIs
At Adobe we have more than 500 engineers and data scientists working on features that use machine learning and AI. In fact, AI is not just the fastest growing skill set, it is also one of Adobe's four innovation drivers. With Adobe Sensei, we want to democratize AI, so that intelligence can be part of every app, every tool, and every experience. The way to make AI scale at Adobe: APIs. We share our lessons learned in weaving AI into all our technology, what it takes to build an API layer for AI, and how to market AI at a Fortune 500 scale.

Speakers
avatar for Lars Trieloff

Lars Trieloff

Principal, Adobe
As a principal at Adobe, Lars’ work spans engineering, product management and marketing. His focus is on combining AI, Serverless computing, and open APIs to enable the next generation of digital experiences. Originally from Berlin, Germany, you find Lars these days mostly in airport... Read More →


Thursday April 5, 2018 13:30 - 14:00 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

13:30 BST

Open Source Machine Learning Deployment
Open source machine learning tools to manipulate and store data and apply complex algorithms to this data have matured greatly in recent years. However, the final step in any successful machine learning project is to put the models into production with exposed APIs as well as monitor, scale and continuously update them. This final stage has generally been built in-house by large companies and can be a considerable risk to the success of projects within more resource constrained companies. This talk will review some of the challenges in deploying machine learning models and introduce some open source projects that attempt to solve these challenges.

Speakers
avatar for Clive Cox

Clive Cox

CTO, Seldon
Clive is CTO of Seldon. Seldon helps enterprises put machine learning into production. Clive developed Seldon's open source Kubernetes based machine learning deployment platform Seldon Core. He is also a core contributor to the Kubeflow and KFServing projects.


Thursday April 5, 2018 13:30 - 14:00 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

14:00 BST

Why the Medical Field is Failing to Keep Up in the A.I. Race
Artificial intelligence is instrumental in changing the way new generations will interact with the world around them. In nearly every field, great new technologies have emerged that benefit society in both practical and personal ways ranging from self-driving cars to smart bot opponents in video games. However, the field that could be reaping the most benefit is also the most reluctant to accept this new wave of technology. The medical field stands to save more lives, spend less public money on data analysis, and free up physician time for more patient interactions among many other benefits. Yet, the most we have seen by way of widespread incorporation of smart technologies in the system was a failed experiment called MYCIN which recommended antibiotics based on bacteria identification. We will focus on the barriers to moving out of the research phase into more practical applications, current research in the field, and future steps needed to advance health technologies.

Speakers
avatar for Anika Mukherjee

Anika Mukherjee

MSc. Health Sciences, University of Ottawa
After graduating with a BSc. Biomedical Sciences in 2016 from the University of Ottawa (distinction: magna cum laude) and interning with medical data science startup company Deski, Anika was well equipped to begin her project developing medical diagnostic tools using machine learning... Read More →


Thursday April 5, 2018 14:00 - 14:30 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

14:00 BST

Predicting the 2018 Oscar Winners with Machine Learning
Is it possible to predict the Oscars? BigML’s Deepnets predicted 6 out of 6 Oscar categories right: best picture, best director, best actress, best actor, best supporting actress, and best supporting actor. But how is it possible to predict a seemingly random event? In this talk, M.Sc. Poul Petersen, Chief Infrastructure Officer at BigML will show how to approach a problem like predicting the Oscars, how to chose the data that is relevant, how to prepare the data, and how BigML's Deepnets work behind the scenes to give the best possible model for your machine learning problem.

Speakers
avatar for Poul Petersen

Poul Petersen

CIO, BigML
Poul Petersen is the Chief Infrastructure Officer at BigML. He has an MS degree in Mathematics as well as BS degrees in Mathematics, Physics and Engineering Physics. With 20 plus years of experience building scalable and fault tolerant systems in data centers, Poul currently enjoys the benefits of programmatic infrastructure, hacking in... Read More →


Thursday April 5, 2018 14:00 - 14:30 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

14:30 BST

Expo - Networking - Tea/Coffee
Thursday April 5, 2018 14:30 - 15:15 BST
Level38 Level38, One Canada Square, London

15:15 BST

AI Startup Pitches
Thursday April 5, 2018 15:15 - 15:30 BST
Level39 - The Sandbox Level39, One Canada Square, London

15:30 BST

Panel: from AI Startup to Exit
Moderators
avatar for John Spindler

John Spindler

CEO, Capital Enterprise
John Spindler is the CEO of Capital Enterprise. John has had over 15 years’ experience as an entrepreneur and business advisor/consultant and as well as being responsible for the day to day management of Capital Enterprise is also a director and co-owner of audio designer and manufacturer... Read More →

Speakers
avatar for Carlos Espinal

Carlos Espinal

Managing Partner, Seedcamp
Mr. Carlos Eduardo Espinal is a Co-Chief Executive Officer, Managing Partner, and Partner at Seedcamp. He focuses on mobile, gaming, social networks, and networking technologies. Prior to that, Mr. Espinal was a Venture Capitalist at Doughty Hanson & Co. Technology Ventures, where... Read More →
avatar for Madeline Parra

Madeline Parra

Product Director, Skyscanner
After five promotions in less than four years, Parra left a successful digital career at GlaxoSmithKline to make the leap into entrepreneurship. She co-founded Twizoo, which gives restaurant and pub recommendations based on what people are tweeting. The start-up has just closed a... Read More →
avatar for Thomas Stone

Thomas Stone

General Partner, AI Seed
Thomas is a Senior Teaching Fellow at UCL and previously Co-founder of PredictionIO, an Open Source Machine Learning Server, acquired by Salesforce.com. Thomas is a PhD graduate in Computer Science and passionate about novel applications of Machine Learning.


Thursday April 5, 2018 15:30 - 16:00 BST
Level39 - The Sandbox Level39, One Canada Square, London

16:00 BST

Microtrends in Artificial Intelligence: The Applied AI Playbook
A ‘micro’ trend is more than a ‘fad’ but less influential than a ‘macro’ trend. Mark Penn, who coined the term, referred to microtrends as “the small forces behind tomorrow's big changes”.

Artificial intelligence, machine learning and blockchain are all macro trends that will persist beyond 10 or 20 years. Microtrends typically last 3 to 5 years before being replaced, or becoming implicit as a result of widespread adoption. An example would be the application of a specific AI technology to a common use case, one which eventually disrupts an entire market horizontal or industry vertical.

Based on 490 interactions with ‘Applied AI’ startups (22% of all Forward Partners' leads in 2017) and examples from our own portfolio, we share our observations to predict the top AI microtrends over the next 5 years.


Speakers
avatar for Chris Corbishley

Chris Corbishley

Investor, Forward Partners
Chris is an Investor at Forward Partners, a leading early-stage VC fund providing startups with a game changing combination of capital + operational support. Their unique approach is helping to build the UK's next generation of talented AI, e-commerce and marketplace businesses. He... Read More →


Thursday April 5, 2018 16:00 - 16:15 BST
Level39 - The Sandbox Level39, One Canada Square, London

16:15 BST

European Machine Intelligence Landscape
Europe deserves a landscape of its own to highlight its talent and expertise. Until recently, its contribution to the innovation and commercialisation of machine intelligence technologies has been under-appreciated. We now see growing self-confidence borne of the success, and continued presence, of local acquired startups like VocalIQ, Swiftkey, Deepmind, Magic Pony Technology, and PredictionIO. London is Europe’s startup centre, mixing capital, proximity to markets, and world-class research hubs. Not to mention that we are proudly European — British, French, and German — with our associated partialities.

Speakers
avatar for Libby Kinsey

Libby Kinsey

Co-founder, Project Juno
Libby began her career at Intellect, the UK’s high tech trade association, and she has an MSci in Mathematics from Imperial College, London. She completed an MSc in machine learning at UCL after 12 years working with technology startups in the UK (including a number with pattern... Read More →
avatar for Dmitry Rastorguev

Dmitry Rastorguev

Quantitative Analyst, Founders Time


Thursday April 5, 2018 16:15 - 16:30 BST
Level39 - The Sandbox Level39, One Canada Square, London

16:30 BST

AI Startup Battle
Find out about the most promising startups in AI and ML: PAPIs hosts the world's 1st startup competition (powered by PreSeries), where participants are judged by an AI on stage—not a human jury!

Apply now for your chance to compete in the Battle: selected startups will be able to present at the conference, exhibit, make connections, and get unique exposure among a highly distinguished audience.

Thursday April 5, 2018 16:30 - 17:00 BST
Level39 - The Sandbox Level39, One Canada Square, London

17:00 BST

Drinks!
Thursday April 5, 2018 17:00 - 19:00 BST
Level38 Level38, One Canada Square, London
 
Friday, April 6
 

08:15 BST

Welcome & Registration
Friday April 6, 2018 08:15 - 09:00 BST
Level39

09:00 BST

The limits of decision making with artificial intelligence
AI systems can fail catastrophically and without warning, a characteristic not welcomed in the corporate environment. Martin will describe the unpredictable nature of artificial intelligence systems and how mastering a handful of engineering principles can mitigate the risk of failure. You’ll learn the kinds of errors artificial intelligence systems make, how to build systems that protect against common errors, and why evaluation can be much harder than it seems.

Speakers
avatar for Martin Goodson

Martin Goodson

Chief Scientist, Evolution AI
Martin Goodson is the chief scientist and CEO of Evolution AI, where he specializes in large-scale statistical computing and natural language processing. Martin has designed data science products that are in use at companies like Time Inc., Hearst, John Lewis, Condé Nast, and Buzzfeed... Read More →


Friday April 6, 2018 09:00 - 09:35 BST
Level39 - The Sandbox Level39, One Canada Square, London

09:35 BST

[Demo] Twilight: Using Generative Tensorial Networks to Illuminate New Chemical Space
Whenever a disease is identified, a new journey into the “chemical space” starts seeking a medicine that could become useful in contending diseases. The journey takes approximately 15 years and costs $2.6bn, and starts with a process to filter millions of molecules to identify the promising hundreds with high potential to become medicines. Around 99% of selected leads fail later in the process due to inaccurate prediction of behaviour and the limited pool from which they were sampled. We are at GTN Ltd addressing the main bottlenecks in drug development by new innovations marring ideas from machine learning and quantum physics ,consequently opening up a new space of possible chemicals unreachable before.

Speakers
avatar for Noor Shaker

Noor Shaker

CEO, GTN Ltd
Prof. Noor Shaker is a co-founder and CEO at GTN Ltd. Before starting GTN, she was an assistant professor at Aalborg University in Copenhagen working on different aspects of machine learning with special interest in generative models. She is the main author of the book “Procedural... Read More →


Friday April 6, 2018 09:35 - 09:45 BST
Level39 - The Sandbox Level39, One Canada Square, London

09:45 BST

[Demo] Asteroid, A Game Engine for Programmers
Asteroid is a game engine especially useful for machine learning practicioners. Unlike other game engines aimed for use in gamedevelopment, which prioritise managing assets like 3D models and animation, Asteroid focuses more on scripting capabilities, allowing users to easily integrate computer graphics with scientific computing and computer vision.

Speakers

Friday April 6, 2018 09:45 - 09:55 BST
Level39 - The Sandbox Level39, One Canada Square, London

09:55 BST

[Demo] NiftyNet: Deep Learning platform for medical image analysis
NiftyNet is a TensorFlow-based open-source convolutional neural networks (CNNs) platform for research in medical image analysis and image-guided therapy. NiftyNet aims to provide many of the tools, functionality and implementations that are essential for medical image analysis but missing from standard general purpose toolkits. Due to its modular structure, NiftyNet makes it easier to share networks and pre-trained models, adapt existing networks to new imaging data, and quickly build solutions to your own image analysis problems. This talk will explore the whys, the whats and the hows of this open source framework.

Speakers
avatar for Jorge Cardoso

Jorge Cardoso

Lecturer in Quantitative Neuroradiology, UCL
I have a BSc in Biomedical Engineering (2006) and an MSc in Medical Electronics and Signal Processing for Biomedical Engineering (2008) from the Universidade do Minho, Portugal, followed by a PhD (2008-2012) and PostDoc (2012-2015) in medical image analysis, machine learning and biomarker... Read More →


Friday April 6, 2018 09:55 - 10:05 BST
Level39 - The Sandbox Level39, One Canada Square, London

10:05 BST

[Demo] Anon AI, Automated Data Anonymisation
Live demo of how Anon AI are using artificial intelligence to automate data anonymisation to make data GDPR compliant.

Speakers
avatar for Harry Keen

Harry Keen

Co-Founder & CEO, Anon AI


Friday April 6, 2018 10:05 - 10:15 BST
Level39 - The Sandbox Level39, One Canada Square, London

10:15 BST

Expo - Networking - Tea/Coffee
Friday April 6, 2018 10:15 - 11:00 BST
Level38 Level38, One Canada Square, London

11:00 BST

Automatic Feature Engineering
As data science evolves as a scientific field, it is time to have a more systematic approach to feature engineering. In this talk, we will present our “human-powered” automatic feature engineering tool. It leverages user’s knowledge to generate expressive features. The goal is to have a versatile, modular and interpretable pipeline that helps data scientists accelerating their workflow.



Du Phan is a data scientist at Dataiku in Paris, where he works in democratising data science. In the past few years, he has been dealing with a variety of machine learning problems, from geospatial analysis to robotics.

Speakers
avatar for Du Phan

Du Phan

Data Scientist, Dataiku
Du Phan is a roboticist turned data scientist at Dataiku in Paris, where he works in democratising data science. In the past few years, he has been dealing with a variety of machine learning problems, from geospatial analysis to robotics. He is a firm believer in transparent, interpretable... Read More →


Friday April 6, 2018 11:00 - 11:30 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

11:00 BST

Tools & Demos
Speakers
avatar for David Arnu

David Arnu

Senior Data Scientist, RapidMiner GmbH
David Arnu works as senior data scientist at RapidMiner. He studied at the University of Dortmund and holds a Master degree in Computer Science. There he worked as a research assistant at the chair of artificial intelligence, later he joined the R&D team of RapidMiner as software... Read More →
avatar for Clive Cox

Clive Cox

CTO, Seldon
Clive is CTO of Seldon. Seldon helps enterprises put machine learning into production. Clive developed Seldon's open source Kubernetes based machine learning deployment platform Seldon Core. He is also a core contributor to the Kubeflow and KFServing projects.
avatar for Alexandre Hubert

Alexandre Hubert

Lead Data Scientist UK, Dataiku
As data scientist, Alexandre has worked on a range of use cases, from creating models that predict fraud to building specific recommendation systems. He especially loves using deep learning with text or sports data. Even when he’s playing sport or having fun with friends, Alexandre... Read More →
avatar for Emanuele Moscato

Emanuele Moscato

Data Scientist, SherlockML
avatar for Jose Antonio Ortega Ruiz

Jose Antonio Ortega Ruiz

CTO, BigML Inc
Jao is part of the founding team of BigML, a startup that applies Machine Learning and other AI techniques to make them accessible to non-specialists. He was hacking for Oblong from 2008 to early 2011. Before that, he worked for Google (from July 2007). From June 2005 to May 2007... Read More →


Friday April 6, 2018 11:00 - 12:00 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

11:30 BST

Smart Data through Automatic Data Search and Extraction
There is an increased focus on deriving business value from data. To exploit the knowledge hidden inside the data, it is necessary to fetch it out of various silos to get a clear picture for the target domain. The challenge is to extract and merge data sets coming from various sources. Our ongoing research project Data Search 4 Data Mining (DS4DM) leads to exploitations of data in graphically designed and highly reusable data mining processes. The goal is to enable an analyst to find and semi-automatically integrate relevant data out of potentially very large corpora.

Speakers
avatar for David Arnu

David Arnu

Senior Data Scientist, RapidMiner GmbH
David Arnu works as senior data scientist at RapidMiner. He studied at the University of Dortmund and holds a Master degree in Computer Science. There he worked as a research assistant at the chair of artificial intelligence, later he joined the R&D team of RapidMiner as software... Read More →


Friday April 6, 2018 11:30 - 12:00 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

12:00 BST

Expo - Networking - Lunch
Friday April 6, 2018 12:00 - 13:30 BST
Level38 Level38, One Canada Square, London

13:30 BST

Unsupervised learning for energy infrastructure inventory
This is a "Data from the trenches" story about optimizing infrastructure inventory in the energy sector by automatically flagging unnecessary field visits. We will explain how to turn an unsupervised problem with messy data into a successful project used on the field. We will discuss challenges we met along the way: working with business experts to create a training set from disparate structured and unstructured data sources; and how to implement unsupervised algorithms in a scalable way. Finally, we will show that messy data can be turned from a hindrance into an opportunity for data scientists.

Speakers
avatar for Alexandre Combessie

Alexandre Combessie

Data Scientist, Dataiku
As a Data Scientist at Dataiku, I design and deploy data projects with Machine Learning (but not only), from prototype to production. Prior to that, I helped build the Data Science team of Capgemini Consulting in France. I started my career as an economist, so I also like interpretable... Read More →


Friday April 6, 2018 13:30 - 14:00 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

13:30 BST

The future of machine learning is decentralized
With the advent of connected devices with computation and storage capabilities, it is now possible to run machine learning workflows entirely on-device. This talk will introduce federated learning and other technologies that enable devices to collaboratively and securely learn ML models, while retaining all data locally. Federated learning improves upon the traditional, fully centralized approaches by reducing the costs and risks related to sensitive data handling, working better in bandwidth and power-constrained environments, and providing a straightforward, effective mechanism for personalization at scale. It also puts users back in control of their data, while still enabling developers to build intelligent applications that leverage insights from that data. Federated learning is already used at-scale by Google - come to this talk to hear how!

Speakers
avatar for Alex Ingerman

Alex Ingerman

Sr Product Manager, Google
Alex Ingerman is a senior product manager at Google, focused on Federated Learning. Before joining Google, he lead the product management team for Amazon Machine Learning. He joined Amazon in 2012, after working on products including web-scale search, content recommendation systems... Read More →


Friday April 6, 2018 13:30 - 14:00 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

13:30 BST

[Workshop] Beyond the API: Advanced ML workflows using DSLs
Predictive APIs provide most of the building blocks to create non-trivial workflows in order to solve real-world problems.  But although they simplify a great deal of the complexity associated with writing ML solutions, workflows based on client-side API scripting lose part of the benefits associated with SaaS solutions (notably, parallelism, re-usability and transparent error handling).  In this workshop we will discuss how to deal with these problems and solve them using a server-side DSL by showing how to implement the well-known Stacked Generalization algorithm using both the BigML Python API and BigML's domain specific language for ML workflows, WhizzML.

Speakers
avatar for Jose Antonio Ortega Ruiz

Jose Antonio Ortega Ruiz

CTO, BigML Inc
Jao is part of the founding team of BigML, a startup that applies Machine Learning and other AI techniques to make them accessible to non-specialists. He was hacking for Oblong from 2008 to early 2011. Before that, he worked for Google (from July 2007). From June 2005 to May 2007... Read More →


Friday April 6, 2018 13:30 - 14:15 BST
Level38 NE Theatre Level38, One Canada Square, London

14:00 BST

Architectures for big scale 2D imagery
I will present research that I conducted during my Ph.D. at University College London and in collaboration with Google. My primary interest lays in the development of neural architectures for 2D imagery problems in big scale. Will present the recently published analysis of different upsampling methods in the decoder part of visual architectures, together with last week ongoing extension for GANs. Will discuss attention mechanism for text recognition and review for what kind of application it can be useful (automatically updating Google Maps based on Google Street View imagery). I will explain the idea behind Inception and what had we change in inception-v3 to have it the best single model on ImageNet 2015 and how does it compare to Resnet architecture which was published 2 weeks after. Together with inception, will present our winning submission to MS COCO 2016 detection challenge and the extensive analysis of different models and backbone architectures inside. At the end will shortly review our UCL effort working with 4096x4096 images at The Digital Mammography DREAM Challenge for breast cancer recognition, where we have achieved 9th among 1375 teams worldwide and 2nd place in the community phase.

Speakers
avatar for Zbigniew Wojna

Zbigniew Wojna

Founder, Tensorflight
Zbigniew Wojna is deep learning researcher and founder of TensorFlight Inc. company providing instant remote commercial property inspection (for risk factors for reinsurance enterprises) based on satellite and street view type imagery. Zbigniew is currently in the final stage of his... Read More →


Friday April 6, 2018 14:00 - 14:30 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

14:00 BST

Techniques for Distributed TensorFlow on Hops
In this talk, we will introduce the latest developments in distributed Deep Learning and how distribution can both massively reduce training time and support parallel experimentation, for large-scale hyperparameter optimization. We will introduce different distributed architectures, including the parameter server and Ring-AllReduce models. In particular, we will describe open-source TensorFlow frameworks that leverage Apache Spark to manage distributed training, such as Yahoo’s TensorflowOnSpark, Databrick’s Deep Learning Pipelines, Uber’s Horovod platform, and Hops’ Hyperparameter Optimization framework for Spark/TensorFLow. We will introduce the different programming models supported and highlight the importance of cluster support for managing GPUs as a resource.

Speakers
avatar for Jim Dowling

Jim Dowling

CEO, Logical Clocks
Jim Dowling is an Associate Professor at KTH Royal Institute of Technology in Stockholm as well as a Senior Researcher at SICS Swedish ICT. He received his Ph.D. in Distributed Systems from Trinity College Dublin (2005) and worked at MySQL AB (2005-2007). He is lead architect of Hops... Read More →


Friday April 6, 2018 14:00 - 14:30 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

14:15 BST

[Tutorial] Rapidminer Auto Model for Product Optimization
In this session we will take a close look at the full capabilities for RapidMiner's Auto Model.

Together we will find the best predictive model to optimize the yield of a production process and explore all the fine tuning options to tweak the outcome. In the spirit of #noblackboxes attendees can take a look under the hood of the model building. We also show how to further increase the applicability of the model by adding extensions.

Speakers
avatar for David Arnu

David Arnu

Senior Data Scientist, RapidMiner GmbH
David Arnu works as senior data scientist at RapidMiner. He studied at the University of Dortmund and holds a Master degree in Computer Science. There he worked as a research assistant at the chair of artificial intelligence, later he joined the R&D team of RapidMiner as software... Read More →


Friday April 6, 2018 14:15 - 15:00 BST
Level38 NE Theatre Level38, One Canada Square, London

14:30 BST

Lessons Learned: An Ethics Roadmap for Real-time Predictions
Ethics is increasingly becoming a buzz word in the world of machine learning. While small and large companies are continuing to innovate their operations with data-driven, predictive algorithms, it remains a challenge for companies to identify what ethical issues are relevant to them and how much risk the issues pose to their business.
In this talk, we share the findings and lessons learned from generating an ethics roadmap at the onset of integrating real-time predictions into the operations of Technical Safety BC, a safety regulator in Canada.

Speakers
avatar for Soyean Kim

Soyean Kim

Head of Research, Technical Safety BC
Soyean Kim is a professional statistician (P.STAT) who is passionate about using data science expertise to contribute to the betterment of society. She currently leads a team of data scientists at Technical Safety BC, a safety regulator in Canada. Her previous leadership roles include... Read More →
avatar for AJung Moon

AJung Moon

CEO & Technical Analyst, Generation R Consulting
Dr. AJung Moon is a CEO and the Technology Analyst of Generation R, a new consultancy that provides internal assessment of algorithmic ethics and robot ethics to help manage risks that are inherent in automation projects. She is also the Director of the Open Roboethics Institute (ORI... Read More →


Friday April 6, 2018 14:30 - 15:00 BST
Level39 - Sandbox 2 Level39, One Canada Square, London

14:30 BST

Automated feature extraction and selection for challenging time-series prediction problems
In the case of predictive modelling, feeding time-series data directly into a machine learning algorithm often leads to sub-optimal performance. Most modern algorithms tend to be slow at learning the embedded time dynamics. This is especially the case in challenging problems such as datasets of small sample size and datasets containing low signal to noise ratio.

A common solution is to include a pre-processing step, namely feature extraction. Given that many features can be extracted from each time-series, this leads to an exponential increase in the dimensionality of the data. Optimal feature set selection can be a time-intensive process and the optimal solution is a function of the choice of algorithm and parameters. The talk will focus on how including automated feature extraction and selection as part of a full machine learning optimisation pipeline can lead to superior results, especially in the case of challenging time-series problems.

Speakers
avatar for Darko Matovski

Darko Matovski

CEO, causaLens
Darko has worked for some of the World’s most prominent hedge funds and research institutions. For example the National Physical Laboratory in London (where Alan Turing worked) and Man Group (a global hedge fund). Darko has a PhD in machine learning and an MBA. He is currently the... Read More →


Friday April 6, 2018 14:30 - 15:00 BST
Level39 - Sandbox 1 Level39, One Canada Square, London

15:00 BST

Expo - Networking - Tea/Coffee
Friday April 6, 2018 15:00 - 15:45 BST
Level38 Level38, One Canada Square, London

15:45 BST

Panel discussion
Moderators
avatar for Alastair Moore

Alastair Moore

Programme Director, UCL School of Management

Speakers
avatar for Libby Kinsey

Libby Kinsey

Co-founder, Project Juno
Libby began her career at Intellect, the UK’s high tech trade association, and she has an MSci in Mathematics from Imperial College, London. She completed an MSc in machine learning at UCL after 12 years working with technology startups in the UK (including a number with pattern... Read More →
avatar for Noor Shaker

Noor Shaker

CEO, GTN Ltd
Prof. Noor Shaker is a co-founder and CEO at GTN Ltd. Before starting GTN, she was an assistant professor at Aalborg University in Copenhagen working on different aspects of machine learning with special interest in generative models. She is the main author of the book “Procedural... Read More →


Friday April 6, 2018 15:45 - 16:15 BST
Level39 - The Sandbox Level39, One Canada Square, London

16:15 BST

Closing Remarks
Alastair Moore (UCL), Florian Douetteau (Dataiku), Louis Dorard (PAPIs)

Friday April 6, 2018 16:15 - 16:30 BST
Level39 - The Sandbox Level39, One Canada Square, London