2020 LMU quantLab Workshop

Official Announcement: www.fm.mathematik.uni-muenchen.de/quantlab/workshops/upcoming/worskhop_2020/index.html

Dates

• 20.02.2020 (AAD, GPU)
• 21.02.2020 (ML)
• 22.02.2020 (optional exercise session)

Location

LMU Munich, Theresienstraße 39, Room B121 (quantLab).

Fee

Industry Practitioner

950 €

Academic

350 €

Note: Students form LMU/TU should visit/register for the lecture Introduction to Machine Learning and Algorithmic Differentiation. This lecture will have a final exam.

Registration

Please contact christian.fries@math.lmu.de.

Presenters

Dr. Benedikt Wilbertz (Machine Learning)

Benedikt Wilbertz is currently Head of Data Science and Machine Learning at Talkwalker, a leading provider of social media analytics solutions. There he is mainly working on deep neural networks and supervised machine learning. He had been prize winner in a Kaggle competition.

Beneath that he is lecturer at Sorbonne Universities Paris and holds a PhD in Probability Theory.

Prof. Dr. Christian Fries (Algorithmic Differentiation)

Christian Fries is head of model development at DZ Bank’s risk control and Professor for Applied Mathematical Finance at Department of Mathematics, LMU Munich.

His current research interests are hybrid interest rate models, Monte Carlo methods, and valuation under funding and counterparty risk. His papers and lecture notes may be downloaded from christian-fries.de/finmath.

He is the author of “Mathematical Finance: Theory, Modeling, Implementation”, Wiley, 2007 and runs finmath.net.

Agenda (Tentative)

Machine Learning

Introduction to Machine Learning

• Concepts of supervised learning
• Bias-Variance trade-off and model performance
• Feature engineering

Linear and non-linear regression models

• Linear models
• Support vector machines

Classification models

• Decision Trees
• Random Forest
• Gradient Boosting
• Model Ensembling

Deep Learning

• Stochastic gradient descent and optimisation for neural networks
• Neural network architectures and applications

Model Interpretability

• Visualisations
• Causal Modeling

Algorithmic Differentiation

Introduction to Algorithmic Differentiation

• Algorithmic Differentiation (AD)
• Adjoint AD

Classic Implementation

• Source Code Transformation
• Operator Overloading

Enabling Software Design Patterns

• Interfaces
• Dependency Injection

Stochastic Algorithmic Differentiation: AAD for Monte- Carlo Simulations

• AD of Conditional Expectations
• AD of Indicator Functions

Application from Finance

• Hedge Simulation
• Margin Valuation Adjustment

Helpful Knowledge

• Basic knowledge of R, Java or another Programming Language.
• Basics in options pricing theory (only for the "Applications from Finance" part)