 |
The Tenth International Conference on Advanced Engineering Computing and Applications in Sciences ADVCOMP 2016 October 9 - 13, 2016 - Venice, Italy |
| Tutorials |
T1. Applied Deep Learning
Prof. Dr. Alexei Dingli, University of Malta, Malta
T2. Data Analytics for Human Resource Management
Prof. Dr. Sandjai Bhulai, Vrije Universiteit Amsterdam, the Netherlands
T3. Security Issues for the IoT: Dealing with Mobile Payments and Secure Element for Objects
Prof. Dr. Pascal Urien, Telecom ParisTech & CNRS UMR 5141, France
Details
T1. Applied Deep Learning
Prof. Dr. Alexei Dingli, University of Malta, Malta
Ancient texts have predicted the arrival of the great flood for centuries. That time is now! We live in a world flooded with big data which flows from smart-phones, watches, vehicles, appliances; almost any piece of technology you can name is designed to communicate back to some online source. The coming decade will see a further increase with the proliferation of the Internet of Things (IOTs). Since it is impossible for a human to process all of that data, we have to make use of Artificial Intelligence (AI) approaches. In recent years, a particular kind of machine learning technique has seen an incredible amount of success across all fields. It won against the international champion of Go, it managed to label images even better than humans in the ImageNet competition and the list of successes can go on. Through this tutorial, we will introduce Deep Learning. Even though the mathematical concepts behind deep learning have been around for decades, software libraries capable of creating and training these networks were only made available in recent years. The power of deep learning is that it gives more flexibility in deciding how to use the data thus removing most of the wild guesses from the equation. Because of this, the decision making process is much more sophisticated resulting in more intelligent machines. Thanks to these advancements, we have self-driving cars, natural language processing is becoming a reality and computer vision is making giant strides into new territories. The tutorial will introduce the participants to Deep Learning using a hands on approach. It will start by giving them an overview, it will explain the various techniques involved and get the participants to try out different applications. In so doing, we hope that the participants can take the examples further, create complex examples and also make use of Deep Learning in their own applications.
T2. Data Analytics for Human Resource Management
Prof. Dr. Sandjai Bhulai, Vrije Universiteit Amsterdam, the Netherlands
Human Resource (HR) management is undergoing a radical change. The US workforce consist of around 160 million workers, and the large expense of most companies is payroll. Most businesses have a payroll that is at least 40% of their total revenue, meaning that the total US payroll expense is many billions of dollars. This raises the question: how well do organizations truly understand what drives performance among their workforce? Do we know why one sales person outperforms his peers? Do we understand why certain leaders thrive and others flame out? Can we accurately predict whether a candidate will really perform well in our organization? The vast majority of hiring, management, promotion, and rewards decisions are made on gut feel, personal experience, and corporate belief systems. However, this is changing due to the availability of data.
T3. Security Issues for the IoT: Dealing with Mobile Payments and Secure Element for Objects
Prof. Dr. Pascal Urien, Telecom ParisTech & CNRS UMR 5141, France
This talk is divided in two parts
Part One: Mobile Payments
This part provides the state of art information needed to understand electronic payment infrastructures dealing with EMV, Paypass, VISA VCPS, or tokenization. Thereafter, we introduce the mobile payment concepts used in systems, such as Google wallet, Apple pay, Android pay or Orange cash. We shortly present the SIMulation project involving Telecom ParisTech and Orange laboratories. The main idea was to deploy secure NFC services in the cloud, secured by SIM modules. From a technological point of view, it is based on Host Card Emulation (HCE), Open Mobile API, RACS servers, and SIM modules providing strong mutual authentication and TLS secure channels. Experiments were successfully performed in France over the 3G/4G networks with commercial mobiles in order to perform NFC mobile payments.
Part Two: Secure Elements for Objects
Tamper resistance, secure communications and storage are consensual requests for emerging Internet of Things (IoT) frameworks. According to the CoAP protocol specified by the IETF, object communications are secured by DTLS or TLS servers. We present the open TLS/DTLS stacks designed for secure elements. The SIM card enabled the secure deployment of billion mobile terminals managed by Mobile Network Operators (MNO). In a similar way, objects could be controlled by dedicated operators. Therefore, TLS/DTLS secure elements could work as identity modules for the IoT. We briefly present the first proof of concept built with Raspberry Pi boards and Javacards.
