September 23-25 | Antwerp Belgium
View More Details  & Register Here

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

Artificial Intelligence (AI) & Machine Learning (ML) [clear filter]
Monday, September 23

15:30 CEST

AI Powered Edge Computing - Sergio Gonzalez, Microsoft
Next-generation services in the Edge, powered by AI capabilities. How hybrid architectures re-key for next-generation killer use cases leveraging 5G and the Edge, by enabling ML/AI capabilities. Experiences and use cases are driven by Microsoft Intelligent Hyperscale Hybrid cloud in different industries (Oil&Gas, Manufacturing, Security, Smart Cities...)

avatar for Sergio Gonzalez

Sergio Gonzalez

Lead Cloud Architect, Microsoft
Once core network and NFV architect, from there I moved to Microsoft as Chief Architect for Telefonica Group. AI and ML can bring the network automation dream and advanced edge computing services come true

Monday September 23, 2019 15:30 - 16:00 CEST
Okapi 2 (Floor 0)
  Artificial Intelligence (AI) & Machine Learning (ML)
  • Experience Level Any
  • Session Slides Included Yes

16:15 CEST

Simplifying Network Automation in 5G Deployments with AI - Shamik Mishra, Altran
5G with its service based architecture has introduced flexibility and agility within its network. However, a 5G network is expected to be heterogeneous with multi-vendor network functions, virtualization, policies, real-time low latency edge computing applications, large density of small micro-data centers etc. This will lead to a complex management model and a potential risk of higher cost of maintenance. On the flip side, the data acquired from these different elements in the network has a potential of simplifying the network slice management, orchestration and performance management. It can improve capacity planning and automation through AI and analytics. In this presentation, we will discuss the challenges and find potential solutions. We will evaluate the role of open source projects like ONAP, gaps in specifications, challenges in data aggregation and monetization.


Shamik Mishra

Asst. Vice President - Research and Innovation, Altran
Shamik Mishra is currently an Asst. Vice President, Research and Innovation at Altran. He is a global lead for the future of networks and compute at Altran group research and innovation. Shamik Mishra has over eighteen years of experience in the Telecom and software industry, providing... Read More →

Monday September 23, 2019 16:15 - 16:45 CEST
Okapi 2 (Floor 0)

17:00 CEST

Bridging the Gap in AI/ML from Cloud to the Edge - Peter Van Ackeren, NXP
Most of the AI/ML learning and model development is being done in the cloud, typically on high-end x86 servers with powerful GPU cards. However, deploying these models to embedded Edge devices which have a much smaller footprint in terms of compute and memory can be a challenge. This session describes how the Edge differs from the cloud in terms of AI/ML processing. It goes on to describe a set of services that can work with different cloud AI/ML frameworks like AWS Sagemaker and Google AutoML, take their output model, convert it into a size and format suitable for embedded Edge devices and deploy over the air.


Peter Van Ackeren

Lead Applications Engineer, NXP

Monday September 23, 2019 17:00 - 17:30 CEST
Okapi 2 (Floor 0)
Tuesday, September 24

11:00 CEST

Network Traffic Analytics through Accelerator and SmartNIC at Edge - Yuming Ma & Mrittika Ganguli, Intel Corporation
Network traffic analytics is critical to data center management and operation efficiency. The regular solution with cloud-based analytics has higher latency and delay for action feedback. This paper proposes an edge-based analytics model with close-loop action feedback to achieve real-time predictive network analytics. Under this solution, the network traffic telemetry data is collected and aggregated by the network devices (e.g. SmartNIC). The AI model is trained at cloud and deployed to the SmartNIC. The predictive analytics and AI inference is managed at the edge device of SmartNIC to achieve real-time close-loop management and mitigation.

avatar for Mrittika Ganguli

Mrittika Ganguli

Ganguli, Intel Corporation
Mrittika Ganguli is a PE and Platform System Architect in Intel’s Network and Custom Logic Group. She has 22 years of experience in platform management, network and storage processing control and data plane, cloud orchestration, telemetry QOS and scheduling Architecture. She has... Read More →
avatar for Yuming Ma

Yuming Ma

Cloud Solution Architect, Intel Corporation
Yuming Ma is a solution architect of Intel DCG (Data Center Group) working on to build cloud solution and system leveraging Intel silicon & adjacency and eco system technology for next generation data center architecture, Prior to Intel, Yuming was chief storage architect of Cisco... Read More →

Tuesday September 24, 2019 11:00 - 11:30 CEST
Okapi 2 (Floor 0)

11:45 CEST

Towards Open Network Intelligence - Faseela K, Ericsson & Frederick Kautz, doc.ai
In this talk, we officially launch the Open Network Intelligence project.

The ONI runner is a Cloud-Native Network Function capable of efficiently training and running AI modules on the data plane. It is designed to ingest L2/L3 data and run in either training or inference mode. The ONI runner communicates with its data plane using memif.

In this talk, we cover ONI's major components and architecture.

avatar for Faseela K

Faseela K

Senior Technical Lead, Ericsson
Faseela is a Technical Lead in the SDN Controller group at Ericsson R&D, and is an active participant of Ericsson contributions to the Opendaylight community.She is a member of the team driving SDN solutions, architecture, design, and product development on the Ericsson Cloud SDN... Read More →
avatar for Frederick Kautz

Frederick Kautz

Head of Edge Infrastructure, Doc.ai
Frederick Kautz is Head of Edge Infrastructure at Doc.ai. He was previously a Principal Software Engineer in the Office of Technology at Red Hat where he focused on improving the overall state of container networking and container+SDN integration. Frederick is an active contributor... Read More →

Tuesday September 24, 2019 11:45 - 12:15 CEST
Okapi 2 (Floor 0)

13:50 CEST

Veo5G Project: A 5G Use Case for Network Data Analytics - Carlos Giraldo & Lorenzo García Cortiñas, Gradiant
This presentation will describe Gradiant's role in the VEO5G R&D project: provide data analytics for networking data in a 5G scenario.

Gradiant researchers will share their experience working with the PNDA.io LFNetworking project and how they manage thousands of Machine Learning models to predict VNF load at the mobile core. VEO5G is an example of the latest technology trends at the service of networking data analytics: Kubernetes, Spark Structured Streaming, and Timeseries Prediction through ML Regression are some of the pieces.


Lorenzo García Cortiñas

Senior Researcher-Engineer | Intelligent Systems Department, Gradiant
Lorenzo García Cortiñas. Telecommunications Engineer by University of Vigo. Joins Gradiant in 2014, working on privacy and security projects on cloud environments (HIGEA, AVATAR, SCAPE); with theese projects he gains expertise on REST API development, SQL and NoSQL databases and... Read More →
avatar for Carlos Giraldo

Carlos Giraldo

Senior Researcher, Gradiant
Carlos Giraldo Rodríguez, 27 November 1984, Vigo, Spain. Carlos graduated from Vigo University in Telecommunications Engineering in 2007 with a final thesis on Delay Tolerant Networks applied to Wireless Sensors. Then he moved to Toulouse to start his PhD on Information Theory and... Read More →

Tuesday September 24, 2019 13:50 - 14:20 CEST
Okapi 2 (Floor 0)

14:35 CEST

Lightning Talk: Automated AI Model Deployment within ONAP - Alassane Samba & Nicolas Edel, Orange
AI will enable more intelligent automation, able to constantly adapt to field reality and specificities.
Through this talk, a prospective vision of the management of an Al model learning, deployment and updating within ONAP will be presented.

It will be accompanied by a demo that shows how new intelligent policies can be learned and updated based on the data collected from ONAP and its managed services at runtime.
Different aspects are considered such as dataset collection through a data lake, model learning from the data lake, model evaluation, model transformation into microservice, microservice deployment, microservice output usage by Policy component and the automation of the whole process facilitating model updating.


Nicolas EDEL

Senior System Administrator, Orange
Nicolas is a FreeBSD evangelist with a strong background in IT and system administration. He worked for HPE and Telisma before joining Orange. He is currently working on Orange ONAP community lab.
avatar for Alassane Samba

Alassane Samba

Research Scientist, Orange
Alassane Samba is a data science specialist. He joined Orange in 2015 as a Research Scientist. His current research interests include artificial intelligence and network management. Since 2017, he has been involved in several Linux Foundation projects such as OPNFV, Acumos and ONAP... Read More →

Tuesday September 24, 2019 14:35 - 14:45 CEST
Okapi 2 (Floor 0)
  Artificial Intelligence (AI) & Machine Learning (ML)
  • Experience Level Beginner
  • Session Slides Included Yes

14:45 CEST

Lightning Talk: Deploy my AI Model using Acumos - Alassane Samba, Orange
In automated Network driven by AI, the quality of AI models will not be the only key differentiator amongst Telcos competitors. The ability to update models and deploy them quickly will be a strong asset. He will give an overview of how Acumos can help Telcos in these tasks.

avatar for Alassane Samba

Alassane Samba

Research Scientist, Orange
Alassane Samba is a data science specialist. He joined Orange in 2015 as a Research Scientist. His current research interests include artificial intelligence and network management. Since 2017, he has been involved in several Linux Foundation projects such as OPNFV, Acumos and ONAP... Read More →

Tuesday September 24, 2019 14:45 - 14:55 CEST
Okapi 2 (Floor 0)

14:55 CEST

Lightning Talk: Change Your Control Loop Logic at Runtime with Adaptive Policy - Liam Fallon, Ericsson
Adaptive Policy is an approach to Policy-Based Management that allows you to specify your application’s logic as adaptable policies that can be deployed and modified as metadata while your application is running. The logic of adaptable policies uses their stimuli and environment to make decisions.

In this lightning talk, Liam gives an overview of Adaptive Policy. He describes how you can use the ONAP Policy Framework to execute and manage your adaptive policies. He sketches how you can evolve your control loop application towards full adaptive policy, where policies adapt to their stimuli and environment at run time.

avatar for Liam Fallon

Liam Fallon

Principal Engineer, Ericsson Software Technology
Dr. Liam Fallon is a Principal Engineer at Ericsson Softare Technology, Ericsson's centre of excellence for open source development. Liam is a network management practitioner and researcher with over 30 years of industrial experience in crafting software systems in that domain. He... Read More →

Tuesday September 24, 2019 14:55 - 15:05 CEST
Okapi 2 (Floor 0)
  Artificial Intelligence (AI) & Machine Learning (ML)
  • Experience Level Beginner
  • Session Slides Included Yes

15:20 CEST

Responsible Use of AI and Blockchain in Telecommunications- Jason Hunt, IBM
AI is having a transformational effect on the operations of the network and surrounding operational and business processes. Blockchain is beginning to make its mark in removing inefficiencies in telco ecosystems. Yet each technology can have issues that can eliminate these business benefits. Machine learning models have been shown to suffer from bias. Some blockchain networks consume massive amounts of energy and leave data in the public. In this talk, we'll discuss the uses of AI and Blockchain in telecommunications, highlight the challenges, and discuss approaches to responsible use in a secured and trusted fashion.

avatar for Jason Hunt

Jason Hunt

Distinguished Engineer, IBM
Jason Hunt is a Distinguished Engineer with IBM. He has over 20 years of experience working with large enterprises to design & develop market-leading solutions in areas such as blockchain, Internet-of-Things (IoT), and cloud networking. Jason also leads IBM’s open source networking... Read More →

Tuesday September 24, 2019 15:20 - 15:50 CEST
Okapi 2 (Floor 0)
  Artificial Intelligence (AI) & Machine Learning (ML)
  • Experience Level Beginner
  • Session Slides Included Yes

16:20 CEST

Lightning Talk: AI Marketplace: We Make AI Accessible and Easy to Use - Jamil Chawki, Orange
Artificial Intelligence is now considered as the main paradigm for the next digital wave and industry 4.0. After an introduction of LF AI foundation projects and landscape, the Lightning Talk will introduce how Orange make AI accessible easy to deploy and to use. By creating a mutualized Open Ecosystem where producers can publish and share AI models that can be easily deployed as a micro-service, Orange AI Marketplace inspires and supports companies in developing their usage of AI, testing and deploying new AI solutions to grow their Business.

avatar for Jamil Chawki

Jamil Chawki

Intrapreneur CEO, Orange
Dr Jamil Chawki leads Orange AI Marketplace at Intrapreneurs Studio. He is Chariman of the Outreach Committee of the Linux Foundation AI Foundation. He was in charge of IT and cloud networks open source activity at Orange and convenor of the Collaborative Teams between ITU-T and ISO/IEC... Read More →

Tuesday September 24, 2019 16:20 - 16:30 CEST
Okapi 2 (Floor 0)

16:40 CEST

Lightning Talk: Can Blockchain and Machine Learning Cooperate? - Lorella Carbone, IBM
The role of collaborative technologies in training AI systems: How to adopt a blockchain approach to contribute to Visual Recognition Framework increasing accuracy in detection of plant diseases. The solution leverages permissioned blockchain like Hyperledger Fabric and Visual Recognition services like IBM Watson Visual Recognition. This new architectural paradigm has been applied in designing and implementing a 5G use case for a major Italian Telco company.

avatar for Lorella Carbone

Lorella Carbone

IT Specialist, IBM
Lorella is one of the technical specialist of the "5G Program" for a major Italian CSP and she joined last month in the IBM 5G Team. In day by day activities, she's able to manage the intersection of many new technologies and application paradigms, providing deep knowledge in AI/ML... Read More →

Tuesday September 24, 2019 16:40 - 16:50 CEST
Okapi 2 (Floor 0)