Event Agenda

Day 1 Cluj: Workshops
2017-11-16
Workshop Room B
Workshop Room A
Day 2 Cluj: Conference
2017-11-17
Conference Stage
Day 3 Sibiu: Conference & Workshops
2017-11-18
Conference
Workshop Room B
Workshop Room A
Conference Stage
Conference
08:30 - 09:30
Registration
08:30 - 09:30
Registration
09:30 - 11:00
Fabian Ottmann and Andrei Dragan (Continental) – How Transmission Control Units Impact the Future of Mobility

Abstract: In this presentation we will talk about the heritage and the market development of transmission control units until today and how Continental’s Transmission BU has to adapt it’s portfolio in order to compete with the challenging and changing marketing of future mobility. This will be expalined by the example of the development of the Pedelec drive units “48V Prime” and “48V Revolution”.

09:30 - 11:00
Mario Alemi (GetJenny) – Build an Open Source Chatbot

Abstract: Using StarChat (starchat.ai), *NIX (Mac or Linux) and Docker, we will have a working chatbot backend. After that, we will setup a simple frontend, running locally, and will go through the main features of StarChat:
– how to read old chat logs and create an answer recommender
– how to create a chatbot using StarChat domain language

Prerequisites: *NIX, Docker, and of course internet connection

11:00 - 11:15
Coffee Break
11:00 - 11:15
Coffee Break
11:15 - 12:45
Florin Baroga (Macadamian) – Build a Voice Enabled Assistant Using Open-source Technologies and Cloud Services

Abstract:IoT (Internet of Things) and Bots are two concepts that dominated the technology world in the last couple of years. They may be still hypes for some people, others may have not heard about them at all, some may have even started successful businesses around them.
Both concepts have been explored intensively by both professionals and hobbyists for a few years. The results are tons of open source (both software and hardware) projects built by enthusiasts but also real products that are being embraced and used by many (in industry, offices or homes): connected sensors, “smart” appliances, assistants like Alexa or Google Home etc. The purpose of this workshop is to demonstrate a use case of both concepts, by creating an internet connected device capable to measure temperature and humidity and send it in cloud for storage and analysis which is also capable to recognize speech input and respond to simple questions or requirements, like “What is the temperature”, “Play music”, etc.

Prerequisites:

OS:
– any should work; preferred: Linux (Debian distro) or Windows

Software:
– Node.js – https://nodejs.org/en/ (preferably latest version)
– Node-RED – https://nodered.org/
– ngrok – https://ngrok.com/
– some light weight code editor (Atom, Visual Studio Code etc) with javascript (Node.js) support
– python3 (optional) – https://www.python.org/downloads/
– mosquitto broker (optional) – https://mosquitto.org/download/
– mplayer (optional) – https://sourceforge.net/projects/mplayerwin/ (windows); for Linux should be found using apt-get (Debian)

Accounts:
– dialogflow account (google account should work) – https://dialogflow.com/
– facebook account + facebook developer account (optional) – https://developers.facebook.com/ – Azure (free) account (optional) – https://azure.microsoft.com/en-us/

Experience:
– some basic javascript experience (main language elements like conditionals, loops, arrays); – Node.js (would be helpful)
– python (would be a plus)

11:15 - 12:45
Artem Pyanykh (Toptal) – Building a Data Warehouse in Google Cloud. DIY Edition

Abstract: Have you wondered about having a Data Warehouse to get the most out of your data, but proprietary solutions were too expensive and constraining while own Hadoop cluster seemed like too much? You don’t need to compromise anymore! From this talk, you’ll learn how to build a simple Data Warehouse and then quickly turn it into a resilient and scalable solution by using the power of Google Cloud.

Prerequisites:

1. Working experience with Python or any other programming language is an absolute must

2. Familiarity with cloud platforms like GCP or AWS will be helpful

3. Familiarity with Data Warehousing concepts is not required as these will be covered during the first part of the session.

12:45 - 14:15
Lunch
12:45 - 14:15
Lunch
14:15 - 15:45
William Oanta (Yopeso) – Generating Faces with GANs in Tensorflow

Abstract: Generative Adversarial Networks (GANs) are one of the hottest topics in present Deep Learning research. An unsupervised technique that Ian Goodfellow introduced in 2014 at Google, GANs continue to advance weekly through new research papers and are expected to give even more impressive results in the coming years. OpenAI, NVIDIA, GoogleBrain and other important AI research centers are presently focused on making new breakthroughs using GANs. As a matter of fact, at the end of the last month, NVIDIA published a new model that is able to generate novel realistic images of unprecedented quality at 1024×1024 resolution. Yann Le Cun, director of AI Research at Facebook and one of the leaders in the Deep Learning community said the following about GANs: “This, and the variations that are now being proposed is the most interesting idea in the last 10 years in ML, in my opinion.”
From inferring novel details from original images, generating realistic next frames in videos, transforming simplistic drawings into photo-realistic sceneries or paintings (and vice-versa), and semi-supervised learning boosting in classification, to generating realistic environments for Reinforcement Learning experiments and generating novel realistic images, GANs play an important role in the quest for Artificial Intelligence by leveraging one of the intelligence’s main assets – imagination. As Albert Einstein answered in an interview with Saturday Evening Post on October 26, 1929: “Imagination is more important than knowledge. Knowledge is limited. Imagination encircles the world.”
Join us for this workshop to build together a face generator in Tensorflow using Generative Adversarial Networks.
Software Stack: Python, Tensorflow, Numpy, Jupyter Notebook, Matplot

14:15 - 15:45
Catalin Huzum (Accenture) – The IoT Platform for everyone – Hands-on with Azure IoT Hub

Abstract: A hands-on demonstration of Azure IoT Hub features with examples in C# but also explanations about general and specific IoT Hub concepts and components.

15:45 - 16:00
Coffee Break
15:45 - 16:00
Coffee Break
15:45 - 16:00
Coffee Break
16:00 - 17:30
Felix Crisan (Netopia) – SQL on Hadoop using Hue, Hive and Impala

Abstract:This workshop is an introduction to ways of doing exploratory analytics on Big Data where the ‘tools of the trade’ are some well known Apache projects like Hive (data warehouse for Hadoop) and Impala (native analytics database for Apache Hadoop), complemented by the open source visual editor from Cloudera – Hue. We’ll use some public datasets and try to explain what kind of queries can be run on HDFS (the storage component of the Hadoop environment) – for batch and interactive queries – and discuss the general performance that can be obtained

16:00 - 17:30
Andreas Theodorou (University of Bath) – Designing and Implementing Transparency in Modular Agents

Abstract: Artificial Intelligence is here, emerging as the defining technology of our time. From personal assistants in our pockets to self-driving cars, it has the potential to transform our lives. It is a new industrial revolution. Yet, we are often encouraged to use AI as a mystical black box, making interactions limited and often uninformative for the end user. This can lead to both safety and privacy concerns; negatively affecting the system’s performance and jeopardizing its functionality. Moreover, such approaches increase the complexity of debugging and tuning intelligent agents, increasing the development time.
In this talk, Andreas will discuss the development of autonomous systems that must deal with changing goals in an unpredictable environment and yet, they remain transparent to inspection. I will present a programming-inspired methodology, Behavior Oriented Design, for implementing modular, behavior-based intelligent systems; such systems range from embodied robots to games AI. Finally, the talk will conclude with a discussion on the benefits of removing the potentially-frightening mystery around “why my robot behaves like that”; helping us to understand, debug, tune, and calibrate our trust towards our machines.

Prerequisites:

-Unity 2017 2.0

-Some knowledge of C# is preferred, but not critical.

-Java 8, at a minimum, installed.

-Familiarity with reactive planning not required, it will be covered at the first part of this workshop

08:00 - 08:50
Registration
08:50 - 09:00
Opening Remarks
09:00 - 09:30
Joanna Gordon (Transform.AI)- The Accelerating AI revolution – Opportunities and Challenges for the Tech Industry
09:30 - 10:00
Alexey Ershov (IBM Watson) – Internet of Things in the Cognitive Era

Abstract: According to IDC, “By 2019, 40% of all digital transformation initiatives – and 100% of all effective IoT efforts – will be supported by cognitive/AI capabilities.” As almost everything — from cars to elevators to washing machines — becomes connected, IoT is changing the way businesses operate. By combining IoT data with cognitive computing, businesses extract valuable insights to improve virtually every aspect of their operations and enable new business models. We will discuss several examples of how IoT can improve operations of a city, a manufacturing plant, a paint shop, an elevator, an offshore gas platform, etc.

10:00 - 10:30
Vlad Lata (KONUX) – Using AI in Predictive Maintenance Solutions – A Practical Application for the Rail Industry

Abstract: Technological advancements in artificial intelligence open up new possibilities for the industry. The railway sector, which is still relying on predominantly manual processes, can gain from digitizing its operations. KONUX offers predictive maintenance solutions for railway infrastructure and combines smart sensors with machine learning algorithms to create a holistic view of assets and their remaining lifetimes. Vlad Lata, CTO of KONUX, gives insights into the practical application of artificial intelligence in a traditional industry.

10:30 - 11:00
Mario Alemi (GetJenny) – Data Analysis from DNA to Artificial Intelligence

Abstract: Mario will address how we can view the evolution of life as being the same as the evolution of data analysis. Organisms learn how to better acquire information about the environment, and therefore extract more energy from it. Showing that human organisations and biological organisms follow similar laws, Mario will demonstrate how data analysis is not just the last hype in marketing, but instead a long-time necessity for business survival.

11:00 - 11:20
Coffee Break
11:20 - 11:50
Sergii Kharagorgiiev (Starship Technologies) – Self-Driving Robots for Revolutionary Local Delivery

Abstract: Starship Technologies delivery robots rely on cameras for localization and obstacle detection, making computer vision critical for successful operation. But there are many challenges for safe and graceful autonomous driving on sidewalks, including real-time processing on embedded hardware with minimal latency, and fusion of visual recognition with other sensor types.

11:50 - 12:20
Kryztof Urban (GumGum) – Large-Scale Automated Text Extraction

Abstract: GumGum performs NLP tasks on millions of websites every month. High quality text extraction (TE) is a key preliminary step for categorization, brand safety analysis, and keyword extraction. Designing a scalable, robust, and language-agnostic solution is a non-trivial task as page layouts differ across domains and may change over time. Our approach involves collecting TE meta-data from human annotators and the Diffbot API. We then apply machine learning algorithms to automatically train and evaluate TE models on a subdomain basis. We will discuss our key findings that lead to optimal performance.

12:20 - 13:00
Fireside Chat: Newton Howard (MIT) & Athanasios Tsanas (University of Oxford) – Using Technological Advances and Machine Learning to Revolutionize Healthcare

Abstract: Recent technological advances provide enormous opportunities to record diverse sets of data. Characteristic paradigms of widely adopted devices are smartphones and smartwatches, and there is growing interest in implantable devices. Contemporary devices provide access to such diverse modalities such as GPS, heart rate, activity and sleep. This talk will highlight some of the latest technological advances, focusing on the latest trends and developments in signal processing and machine learning to capitalize on the raw data and make key breakthroughs in healthcare applications.

13:00 - 14:00
Lunch
14:00 - 14:30
Catalina Butnaru (City.ai) – Working with AI – Future-proofing Your Development Process

Abstract: Catalina is currently developing a methodology that engages employees in creating an AI solution that fits the business’s core business model, culture and processes, without causing unnecessary job displacement.

She will talk about compatible processes for the successful delivery of AI-first solutions, with less risk and increased business impact, while facilitating the engagement of key people from the entire organisation.

14:30 - 15:00
Robin Hayden (Paddy Power Betfair) – Machine Learning Adoption Challenges and Strategies

Abstract: Machine learning represents a sea change in computing and an enormous opportunity to super charge every business process with automated intelligence. Although artificial intelligence has already started surfacing in main stream media and has been proven to work in a wide range of applications, adoption rates in most established businesses are still fairly low.

This talk is about getting ahead of that adoption curve by identifying common obstacles to adoption, plausible strategies and tactics to overcome them and gradually building the skills, platforms and familiarity needed to grasp the opportunity while it still has the potential to be a significant differentiator.

15:00 - 15:30
Berthold Arweiler (Continental) – Industry 4.0 – How Will Digital Transformation and Smart Factory Influence Manufacturing

Abstract: The Digital Transformation is influencing most of the areas in our life. Data is the gold of the 21th century and Internet companies earn a lot of money with our data in providing free of charge services and they were and they are still improving the capability to make even more use out of these data.

How will these technologies influence manufacturing and which technology areas have potential to support the vision of a Smart Factory where data driven decisions will replace or at least support the todays decision processes.

15:30 - 16:00
Fabien Benetou (LucidWeb) – Making Data Tangible: Using VR for Data Visualization

Abstract: Data is all around us, abstract, used in models, computations… so important yet so hard to
grasp. Virtual reality (VR) is making us literally stand up to see, grab and manipulate data. What has been so far mostly science fiction illustrations becomes now the most efficient way to interact with information, involving our entire body and mind to better make sense
of data.

The talk will introduce the fundamentals of VR, how the hardware is able to track our position and orientation in space, then how software stack enables the recent tremendous progress in tracking and display. Building a VR experience isn’t sufficient though, delivery matters, especially as user expect nearly frictionless content delivery. Specific dive in WebVR, how VR can be delivered directly and instantly with the web, will be showcase. A live coding session will explore a dataset to bring everything together as 1 pragmatic example.

16:00 - 16:20
Coffee Break
16:20 - 16:50
Virginia Dignum (Delft University of Technology) – Design and Evaluation of Human-agent Teamwork

Abstract: Researchers and developers are pursuing increasingly sophisticated roles for autonomous systems. Whether working within networked systems as software agents or embedded in robots and unmanned vehicles, what makes these systems valuable is their intelligent, active, and adaptive nature. These qualities are often characterized in intelligent systems literature by the word “autonomy”-a catch-all label that highlights the qualities of self-directedness and self-sufficiency in task performance. Human-agent-robot-teamwork (HART) research seeks to bring together the best thinking from diverse research communities in order to advance current and anticipated applications of intelligent human-machine collaboration, including the participation of humans as first-class citizens in collaboration with autonomous systems. This would enable autonomous systems not merely to do things for people, but also to work together with people and other systems-the inevitable next leap-forward required in autonomous system design and deployment.

16:50 - 17:20
Shahar Nechmad (Credit Stacks) – Chatbots and Voice Based Interfaces: Where We Are and Where We Need to Get To

Abstract: It seems that all we hear about these days is the chatbot wars. The big companies are spending billions building new voice activated interfaces while 80% of businesses are expected to have their own chatbot by 2020. At the same time, most of us are still not using voice based UI on a daily basis and most experiences out there tend to be underwhelming.
In this talk we will explore where voice based services are currently today, what they are doing wrong and what we as developers and entrepreneurs need to do to take them to the next level, and prepare our businesses for the future.

17:20 - 17:50
Camilo Martinez (Booking.com) – The Lifecycle of a Machine Learning Product

Abstract: Visitor Profiling teams at Booking.com have been developing Machine Learning and Data Science based products for several years now. Camilo will be talking about how to approach developing these products to bring value to your customers, as well as giving a brief summary of the technologies used at each step of the lifecycle.

17:50 - 18:00
Codiax Closing Ceremony
08:30 - 09:00
Registration
09:00 - 09:10
Opening Remarks
09:10 - 09:45
Fireside Chat: Newton Howard (MIT) & Athanasios Tsanas (University of Oxford) – Using Technological Advances and Machine Learning to Revolutionize Healthcare

Abstract: Recent technological advances provide enormous opportunities to record diverse sets of data. Characteristic paradigms of widely adopted devices are smartphones and smartwatches, and there is growing interest in implantable devices. Contemporary devices provide access to such diverse modalities such as GPS, heart rate, activity and sleep. This talk will highlight some of the latest technological advances, focusing on the latest trends and developments in signal processing and machine learning to capitalize on the raw data and make key breakthroughs in healthcare applications.

09:45 - 10:10
Dinu-Tudor Gruian – An introduction to Continental’s approach for manufacturing test technologies

Abstract: Continental must meet the challenge of standardising production test systems across the globe by having a future-proof test platform while preserving the fragile balance between the fixed standard core, flexibility, and scalability.
The business impact of SW and SW development for manufacturing is also an important part of the success recipe for a worldwide manufacturing organization. In this presentation you will hear about Continental’s standardisation model for testing in production and how SW is impacting today’s automotive environment.

10:10 - 10:35
Alexey Ershov (IBM Watson) – Internet of Things in the Cognitive Era

Abstract: According to IDC, “By 2019, 40% of all digital transformation initiatives – and 100% of all effective IoT efforts – will be supported by cognitive/AI capabilities.” As almost everything — from cars to elevators to washing machines — becomes connected, IoT is changing the way businesses operate. By combining IoT data with cognitive computing, businesses extract valuable insights to improve virtually every aspect of their operations and enable new business models. We will discuss several examples of how IoT can improve operations of a city, a manufacturing plant, a paint shop, an elevator, an offshore gas platform, etc.

10:35 - 10:50
Coffee Break
10:50 - 11:15
Sergii Kharagorgiiev (Starship Technologies) – Self-Driving Robots for Revolutionary Local Delivery

Abstract: Starship Technologies delivery robots rely on cameras for localization and obstacle detection, making computer vision critical for successful operation. But there are many challenges for safe and graceful autonomous driving on sidewalks, including real-time processing on embedded hardware with minimal latency, and fusion of visual recognition with other sensor types

11:15 - 11:40
Kryztof Urban (GumGum) – Large-Scale Automated Text Extraction

Abstract: GumGum performs NLP tasks on millions of websites every month. High quality text extraction (TE) is a key preliminary step for categorization, brand safety analysis, and keyword extraction. Designing a scalable, robust, and language-agnostic solution is a non-trivial task as page layouts differ across domains and may change over time. Our approach involves collecting TE meta-data from human annotators and the Diffbot API. We then apply machine learning algorithms to automatically train and evaluate TE models on a subdomain basis. We will discuss our key findings that lead to optimal performance.

11:40 - 13:00
Mario Alemi (GetJenny) – Build an Open Source Chatbot

Using StarChat (starchat.ai), *NIX (Mac or Linux) and Docker, we will have a working chatbot backend. After that, we will setup a simple frontend, running locally, and will go through the main features of StarChat: – how to read old chat logs and create an answer recommender – how to create a chatbot using StarChat domain language

11:40 - 13:00
Mihai Schiopu (Continental)- From ADAS to 2025AD

Abstract: Despite the promising technological advances, automated driving still faces challenges on the road ahead. One key point: Acceptance. Self-driving cars will have to be programmed to make ethically acceptable decisions. We will go through existing ADAS systems, present a vehicle equipped with Continental sensors and discuss about challenges coming with self-driving cars.

13:00 - 14:30
Lunch & Networking
Select date to see events.