2022
RAISE CoE Training: LSTM and GRU
Morris Riedel, Reza Hassanian
2022-10-31, Public YouTube Online Training and Teaching Seminar
Abstract
The RAISE CoE EU project works on nine different engineering use cases (see: https://www.coe-raise.eu/use-cases) that bring researchers and experts from the industry together to co-design intertwined AI and HPC methodologies for the Exascale era. These methodologies’ goal is to be usable by various scientific and engineering applications to enable AI at Exascale (see: https://www.coe-raise.eu/ai-exascale). While several of these applications adopt deep learning methods for image processing, such as Convolutional Neural Networks (CNNs), many other use cases require sequence analysis methods for time series analysis. That is particularly the case for Computational Fluid Dynamics (CFD) domain applications. This seminar informs about the general approach of Recurrent Neural Networks (RNNs) for sequence analysis. Then, building on that foundation, a brief introduction to Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs) explain cutting-edge deep learning methods. Finally, the seminar further outlines the initial lessons learned by CoE RAISE in adopting LSTMs and GRUs in CFD applications. That includes comparisons of performance and the use of HPC resources.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~2.80 MB (pdf) ]
[ CoE RAISE Relevance of LSTM and GRU Models in CoE RAISE ~5.27 MB (pdf) ]
[ CoE RAISE Introduction to LSTM and GRU Models ~8.38 MB (pdf) ]
[ Thanks Slides ~1.36 MB (pdf) ]
Video in post production, to appear
RAISE CoE Training: Transformer Models
Morris Riedel, Christian Lessig
2022-09-26, Public YouTube Online Training and Teaching Seminar
Abstract
The CoE RAISE project develops several AI methods in nine compute-intensive and data-intensive use cases. The use case researchers leverage various AI methods using heterogeneous high-performance computing (HPC) systems and co-design the RAISE unique AI framework towards Exascale. After a short introduction to CoE RAISE, the seminar demonstrates how CoE RAISE and other computational-intensive and data-intensive communities can benefit from cutting-edge AI approaches such as transformer models. Furthermore, it will introduce the benefits of representation learning for unsupervised learning in particular and attention mechanisms in transformer models in general. Finally, application examples and use cases are part of the seminar in the specific context of the different models.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~2.79 MB (pdf) ]
[ CoE RAISE Relevance of Transformer Models in CoE RAISE ~4.67 MB (pdf) ]
[ CoE RAISE Introduction to Transformer Models ~6.50 MB (pdf) ]
[ Thanks Slides ~1.73 MB (pdf) ]
Video in post production, to appear
RAISE CoE Training: Towards a CoE RAISE Unique AI Software Framework for Exascale
Morris Riedel, Rakesh Sarma, Marcel Aach
2022-08-29, Public YouTube Online Training and Teaching Seminar
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~2.83 MB (pdf) ]
[ CoE RAISE Relevance of AI Software Framework in CoE RAISE ~4.66 MB (pdf) ]
[ CoE RAISE Unique AI Framework Overview ~5.78 MB (pdf) ]
[ Thanks Slides ~1.89 MB (pdf) ]
Video in post production, to appear
RAISE CoE Training: Using OpenML for sharing datasets, algorithms, and experiments
Morris Riedel, Joaquin Vanschoren
2022-05-31, Public YouTube Online Training and Teaching Seminar
Abstract
The CoE RAISE project develops a unique AI framework leveraging high-performance computing (HPC) environments to enable faster machine and deep learning model training. That benefits the CoE RAISE application use cases and many other scientific and engineering domains that adopt AI techniques. This seminar will introduce the OpenML platform as one option of CoE RAISE to link to the larger AI community and offer components from the unique AI framework to many researchers worldwide using OpenML. The seminar describes the approach of OpenML as an open platform for sharing machine learning datasets, algorithms, and experiments. In addition, the seminar discusses potential collaboration opportunities through interoperability between OpenML and CoE RAISE. Hence, the OpenML platform acts as one outreach channel to the international AI community, potentially leveraging open standards such as the Open Neural Network Exchange (ONNX).
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~2.84 MB (pdf) ]
[ CoE RAISE Relevance of OpenML in CoE RAISE ~9.46 MB (pdf) ]
[ Thanks Slides ~2.29 MB (pdf) ]
Video in post production, to appear
RAISE CoE Training: Quantum Support Vector Machine Algorithms
Morris Riedel, Amer Delilbasic, Edoardo Passeto
2022-04-21, Public YouTube Online Training and Teaching Seminar
Abstract
Today, all application use cases of the CoE RAISE project leverage the power of high-performance computing (HPC) systems via Central Processing Units (CPUs) or Graphical Processing Units (GPUs). However, several use cases have recently started exploring a disruptive computing technology known as quantum computing. This seminar will introduce one particular quantum computing approach called quantum annealing and will describe why it may be one form of computing in the future. Researchers use quantum annealing in CoE RAISE to solve complex optimization problems inherent in machine and deep learning algorithms. The seminar will provide particular examples in the context of a traditional but still relevant machine learning model called support vector machines (SVMs). Besides showing classification examples with SVMs, the seminar will also cover support vector regression techniques.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~2.84 MB (pdf) ]
[ CoE RAISE Relevance of Quantum Computing in CoE RAISE ~9.41 MB (pdf) ]
[ Thanks Slides ~2.70 MB (pdf) ]
Video in post production, to appear
RAISE CoE Training: Graph Neural Networks
Morris Riedel, Leo Nicoletti, Eric Wulff
2022-03-31, Public YouTube Online Training and Teaching Seminar
Abstract
The application use cases of the CoE RAISE project co-design and use a unique AI framework to develop novel AI models and techniques that benefit from high-performance computing (HPC). Most use cases develop models in innovative deep neural networks, such as Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), or AutoEncoders. This seminar introduces the idea of Graph Neural Networks (GNNs) and their benefits in identifying patterns in graphs with complex relationships and interdependencies of particular objects. GNNs are helpful because many data in applications have an underlying graph structure and a non-regularity in their data structures. While the seminar will cover several fundamentals about GNNs, it will also include complex use case application examples about using GNNs in numerical simulations and particle reconstruction.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~2.87 MB (pdf) ]
[ CoE RAISE Relevance of Graph Neural Networks in CoE RAISE ~7.68 MB (pdf) ]
[ Short Introduction to Graph Neural Networks ~2.66 MB (pdf) ]
[ Thanks Slides ~1.66 MB (pdf) ]
Video in post production, to appear
2021
RAISE CoE Seminar: Accelerating ML with GraphCore
Morris Riedel, Pawel Gepner, Alexander Titterton
2021-11-23, Public YouTube Online Training and Teaching Seminar
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~3.73 MB (pdf) ]
[ CoE RAISE Relevance of GraphCore IPUs in CoE RAISE ~7.68 MB (pdf) ]
[ Thanks Slides ~1.88 MB (pdf) ]
Video in post production, to appear
RAISE CoE Seminar: Hyperparameter Tuning with Ray Tune
Morris Riedel, Marcel Aach
2021-10-29, Public YouTube Online Training and Teaching Seminar
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~3.71 MB (pdf) ]
[ CoE RAISE Relevance of Hyperparameter Tuning in CoE RAISE ~8.40 MB (pdf) ]
[ Introduction to Hyperparameter Tuning and Neural Architecture Search ~12.4 MB (pdf) ]
[ Thanks Slides ~2.02 MB (pdf) ]
Video in post production, to appear
RAISE CoE Seminar: MLOps with ClearML
Morris Riedel, Kurt de Grave
2021-09-30, Public YouTube Online Training and Teaching Seminar
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~3.66 MB (pdf) ]
[ CoE RAISE Role of MLOps ~8.56 MB (pdf) ]
[ Introduction to MLOps with ClearML ~6.52 MB (pdf) ]
[ Thanks Slides ~1.99 MB (pdf) ]
Video in post production, to appear
RAISE CoE Seminar: Brief Introduction to Autoencoders
Morris Riedel, Rakesh Sarma
2021-08-31, Public YouTube Online Training and Teaching Seminar
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~3.69 MB (pdf) ]
[ CoE RAISE Role of Autoencoders ~6.34 MB (pdf) ]
[ Autoencoders – A Brief Introduction and Overview ~8.29 MB (pdf) ]
[ Thanks Slides ~1.99 MB (pdf) ]
Video in post production, to appear
RAISE CoE Seminar: Distributed Deep Learning
Morris Riedel, Rocco Sedona, Marc Sergent, Marcel Aach
2021-07-29, Public YouTube Online Training and Teaching Seminar
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~3.71 MB (pdf) ]
[ CoE RAISE Need for Distributed Deep Learning Slides ~9.78 MB (pdf) ]
[ Thanks Slides ~2.13 MB (pdf) ]
Video in post production, to appear
RAISE CoE Seminar: High Performance Data Analytics with the Helmholtz Analytics Toolkit (HeAT)
Morris Riedel, Claudia Comito, Charlotte Debus
2021-06-28, Public YouTube Online Training and Teaching Seminar
Abstract
The CoE RAISE project develops many AI methods in nine compute-intensive and data-intensive use cases. The use case researchers leverage various AI tools on heterogeneous high-performance computing (HPC) systems and co-design the RAISE unique AI framework towards Exascale. The seminar demonstrates how CoE RAISE and other computational-intensive and data-intensive communities can benefit from the free Helmholtz Analytics Toolkit (HeAT). The goal of HeAT is to fill the gap between data analytics and machine learning libraries with a strong focus on single-node performance on the one hand and traditional HPC on the other. HeAT’s generic Python-first programming interface integrates seamlessly with the existing data science ecosystem in CoE RAISE. It makes it as effortless as using NumPy to write scalable scientific and data science applications. The seminar provides a sophisticated introduction to Heat and its use cases and discusses a possible adoption of HeAT in the RAISE unique AI framework design.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~3.74 MB (pdf) ]
[ CoE RAISE Towards Unique AI Framework Methodologies Slides ~7.94 MB (pdf) ]
[ Thanks Slides ~2.57 MB (pdf) ]
Video in post production, to appear
RAISE CoE Seminar: Git-based Data Management with the Open-Source DataLad Tool
Morris Riedel, Michael Hanke, Kaustubh Patil
2021-05-28, Public YouTube Online Training and Teaching Seminar
Abstract
The seminar demonstrates how CoE RAISE and other computational-intensive and data-intensive communities can benefit from the free DataLad tool. It enables researchers to discover data since it has built-in support for metadata extraction and search. HPC & AI researchers often consume data in different ways requiring direct access to individual files, especially when using a few files from some large datasets for analysis. DataLad enables that and supports also sharing datasets with the public or just some colleagues on platforms without the need for a central service for publishing datasets. Version control systems such as GIT are a de-facto standard for open-source software development. A similar level of tooling enables the DataLad tool for data management and analysis. HPC & AI researchers benefit from comprehensively track the exact state of any analysis inputs that produced results across the entire lifetime of a project and multiple datasets, enabling reproducibility.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~4.41 MB (pdf) ]
[ CoE RAISE Use Case Foundations and Requirements for Data Management Slides ~7.84 MB (pdf) ]
[ Thanks Slides ~2.54 MB (pdf) ]
RAISE CoE Online Seminar: HPC Systems Engineering in the Interaction Room
Morris Riedel, Matthias Book, Helmut Neukirchen
2021-04-08, Public YouTube Online Training and Teaching Seminar
Abstract
The recently started RAISE CoE EU project works on nine different engineering use cases (see: https://www.coe-raise.eu/use-cases) that bring researchers and experts from the industry together to co-design intertwined AI and HPC methodologies for the Exascale era. These methodologies’ goal is to be usable by a wide variety of scientific and engineering applications to enable AI at Exascale (see: https://www.coe-raise.eu/ai-exascale).
One key element of the above CoE project co-design process is to perform HPC systems engineering in the Interaction Room because the design, development, and deployment of scientific computing applications are very complex holistically representing a complicated system. It requires scientific, industry, HPC, AI, and excellent software engineering expertise to enable scalability for Exascale. The cooperation and communication between experts from these quite different disciplines can be difficult.
This seminar informs about the general Interaction Room technique that facilitates interdisciplinary collaboration in complex software projects, emphasizing intertwined AI and HPC. An Interaction Room is a (physical or virtual) room that is outfitted with several large analogue or digital whiteboards known as canvases. They are used to visualize and facilitate discussion of critical aspects of a complex software system. Each canvas is dedicated to modeling a particular perspective on the system. The key difference to other modelling techniques is that models in the Interaction Room are kept deliberately informal. Hence, the goal is not to create a perfect specification but to encourage stakeholders from diverse backgrounds to discuss those aspects that are essential to the software project’s success.
The seminar demonstrates how the CoE RAISE aims to perform co-design with this Interaction Room technique to understand the domain requirements, understand technical restrictions, identify aspects of particular scientific value, and identify the most critical risks of those projects. The seminar further outlines initial lessons learned in intertwined HPC and AI applications. Using the Interaction Room at an early project stage helps prevent costly misunderstandings and oversights later on and has already proven helpful in numerous complex information systems projects.
[ Video in RAISE Public YouTube Channel ]
[ Welcome Slides ~5.14 MB (pdf) ]
[ CoE RAISE Use Case Foundations and Lessons Learned from Fact Sheets Slides ~8.28 MB (pdf) ]
[ Thanks Slides ~3.37 MB (pdf) ]
Talk at IPDPS HCW 2021: Practice and Experience in Using Parallel and Scalable Machine Learning with Heterogenous Modular Supercomputing Architectures
Morris Riedel
Heterogenity in Computing Workshop (HCW) held in conjunction of the
35th IEEE International Parallel and Distributed Processing Symposium (IPDPS)
Portland, Virtual Conference,
2021-05-17 – 2020-05-21
[ EVENT ] [ Slides ~21.3 MB (pdf) ]
Short Introduction to DataLad and Juelich Activities
Morris Riedel
2021-02-03, Public YouTube Online Training and Teaching Seminar, Public Helmholtz International BigBrain Analytics and Learning Laboratory (HIBALL 2021) Winterschool for Online Learning, BigBrain Data and Tools February 3 – 4, 2021, Virtual with YouTube Lectures
[ Event ] [ Slides ~4.81 MB (pdf) ]
2020
Chairing Euro-Par 2020 Best Paper Session
Morris Riedel
Topic Chair Data Management, Analytics and Deep Learning
26th Euro-Par Conference, August 24-28, 2020, Warsaw, Poland
[ Event ]
Chairing Euro-Par 2020 Topic 05 Data Management – Euro-Par 2020 Session
Morris Riedel
Topic Chair Data Management, Analytics and Deep Learning
26th Euro-Par Conference, August 24-28, 2020, Warsaw, Poland
[ Event ]
UTmessan 2020 – Demystifying Quantum Computing
Morris Riedel
UTmessan 2020, February 7, 2020, Harpa, Reykjavik, Iceland
[ Event ]
2017
Invited Tutorial – Deep Learning using a Convolutional Neural Network
Morris Riedel
Ghent University, six lectures including exercises
2017-11-30 – 2017-12-01
[ Event ]
Invited Tutorial – Introduction to Machine Learning Algorithms
Morris Riedel
Ghent University, six lectures including exercises
2017-11-23 – 2017-11-24
[ Event ]
PRACE 2017 Spring School – Introduction to Parallel and Scalable Machine Learning – Basics
Morris Riedel
The Cyprus Institute, April, 25 – 27, 2017, Nicosia, Cyprus
[ Event ]
PRACE 2017 Spring School – Introduction to Parallel and Scalable Machine Learning – Parallelization Benefits
Morris Riedel
The Cyprus Institute, April, 25 – 27, 2017, Nicosia, Cyprus
[ Event ]
2016
UTmessan 2016 – Societal Impact of High Performance Computing in Science and Engineering
Morris Riedel
UTmessan 2016, February 5 – 6, 2016, Harpa, Reykjavik, Iceland
[ Event ]
2015
PRACE XSEDE Interoperability Projects – Smart Data Analytics for Earth Sciences across XSEDE and PRACE
Morris Riedel
ECSS Online Webinar, March 6, 2015
(Talk starts at 32 minutes 11 seconds)
[ Event ]
2011
Interview: Grid Interoperation Now GIN OGF Working Group at OGF31 as GIN CG Co Chair
Morris Riedel
Open Grid Forum (OGF) 31 Conference, March 21, 2011, Taipei, Taiwan
[ Event ]
2009
Interview: Production Grid Infrastructure PGI OGF Working Group at OGF25 as PGI WG Co Chair
Morris Riedel
Open Grid Forum (OGF) 25 Conference, March 2 – 3, 2009, Catania, Italy
[ Event ]