Interactive Workshops - Monday March 18, 2019


As many of the best minds in AI are in NYC for the main conference, we have taken the opportunity to put together a unique workshop day which is separate from the main conference and is aimed at analysts, quants and researchers from asset managers who want to learn new AI coding methods and the latest theory around applications for ML /DL / NN.

At these in-depth, “roll up your sleeves” sessions, you will receive expert, practical, hands-on tuition on the very latest tools and techniques you can utilize every day at work. A series of parallel sessions allow you to choose the content most useful to you.
These sessions may not be suitable for beginners. Knowledge of programming and some understanding of deep learning is a necessity, and a list of requirements will be provided for each session, allowing you to ensure your suitability, and to download any necessary software in advance. You are a data scientist, software developer, data engineer, or financial data analyst who wants to use the very latest tools, technology and techniques to: analyze data sets; optimize portfolio risk models; refine trading strategies & back test. By attending these workshops, you will stay at the cutting-edge of your field. A great opportunity for personal development,
as well as ensuring that your firm keeps its edge.

Topics that will be covered:
• Computational ML & AI using Julia, Python & R
• Deep learning
• Neural networks
• Regression and machine learning techniques
• Portfolio simulation & backtesting strategies

Details on the workshops are listed below, please note that attendees will be required to bring their own laptops and to download application specific programs in advance to allow, for example, real ML queries to be created using real data on the latest cloud based GPU processors.

The workshop and conference are available to book separately. Workshops include lunch and refreshments, plus entry to the main conference Ice Breaker Reception, where you can mingle with speakers and delegates to the main conference on the evening of Monday March 18.

TIDY TRADING WITH R: An Introduction to R, Machine Learning and Data Visualization for Quants (Full Day)

R is a fast growing language in the world of finance, due largely to its packages for portfolio analysis, data visualization and machine learning. In this workshop, we will build a simple but real trading strategy from beginning to end. We will start by importing, wrangling and exploring raw data, and then progress to constructing and testing signals with machine learning, implementing a strategy and backtesting results. Finally, we will build an interactive dashboard to communicate the results of our work. In the process, we will explore and become familiar with packages and code that can serve as a template for further work or applications. We will emphasize clean, reproducible code and data visualization as much as we emphasize the tools of modeling and machine learning so that attendees leave with the ability to run machine learning models, but also communicate the results in a compelling way.

- learn to import data via databases, APIs and excel
- explore tools for cleaning and wrangling data
- delve into R's data visualization capabilities with ggplot, plotly and highcharter
- build interactive dashboards for data exploration with Shiny
- use machine learning to model financial data and implement a strategy with the tidy models framework
- visualize and communicate the results of our machine learning work with lime and Shiny

Who should attend
- The course is aimed at people with financial industry knowledge or experience who want to learn R.
- Excel users
- portfolio managers
- market analysts and researchers
- quants and aspiring quants
- CFA charterholders

- No coding experience is necessary, but a desire to spend a day learning and digging into R code is necessary. 


(Full Day)

Course Objectives
1. Introduction to various machine learning and artificial intelligence concepts
2. Learn machine learning, deep learning and AI concepts 
3. Provide training so that attendees can start writing applications in AI
4. Provide ability to run real machine learning production examples
5. Understand programming techniques that underlie the production software

The concepts will be taught in Julia, the fastest and most productive modern high-level language for numerical computing and machine learning - but can be applied in
any language with which the audience is familiar.

The audience are expected to have learnt typical first 2 years of undergrad mathematics and linear algebra and have exposure at least one programming language.
Professionals with a few years of experience in data science will also benefit from this course.

Course outline
1. Representing Data with Models: Use of functions and parametric functions to build models
2. Model Complexity: What is learning from a computational point of view? How does a computerlearn?
3. Exploring Data with Unsupervised Learning: Dimensionality reduction for image classification
4. Applications Using Unsupervised Machine Learning
5. Introduction to Supervised Machine Learning
6. Practical Applications using Supervised Machine Learning, for example object detection
7. Introduction to Neurons: Learning with a single neuron
8. Introduction to Flux.jl: Learning with a single neuron using Flux.jl
9. Introduction to Neural Networks: Building single layer neural nets with Flux.jl
10. Introduction to Deep Learning: Multi-layer neural networks with Flux.jl
11. Handwriting Recognition with Neural Networks

(Full day)

Finance Practitioners and Machine Learners will learn ML techniques in Finance and Implementation of ML projects in Finance. We will cover the most relevant ML and AI Algorithms. An excellent blend of mathematics, financial intuition and Python to learn Machine and Artificial Intelligence in Finance. 

Quantitative Finance 
- Review quantitative finance 
- Alternative data 
Machine Learning Modelling 
- Mathematics of machine learning 
- Machine learning modelling framework
Supervised Learning: Classification
- Logistic regression and Softmax regression 
- SVM’s and CART’s 
- Boosting and bagging: Random facts
- AdaBoost + XG Boost
Supervised Learning: Regression 
- Modern linear regression 
- Non-Linear regression
- Neural networks
- Deep neural networks
Supervised Learning: Deep Learning 
- Mathematics of deep learning 
- Deep learning architectures
Reinforcement Learning Natural Language Processing 
- Sentiment analysis – NLTK
Python and Exercises



Big data and AI represent an opportunity to impact the bottom line of organizations. By building advanced analytics and machine learning models on top of these large repositories of data, business decisions can be vastly improved. Some tasks that stand to gain the most from such improvements include: predicting fraud, determining trading opportunities, improving the customer experience/journey, and even forecasting which employees are at risk for switching companies.However, implementation at scale often is a challenge in and of itself. For example, how do you import a distributed system of data or a repository of files of various sizes/types? Is it tedious to bring models to the data and deploy analytics directly on Hadoop or Spark clusters? How much recoding or refactoring is necessary?

In this hands-on session, we will build AI models with MATLAB through various machine learning and deep learning examples. As we move from prototype to production we will then evaluate and optimize model performance, scale to big data systems such as Spark and Hadoop, and rapidly deploy your predictive models into production.

• Learn the fundamentals of deep learning and traditional supervised and unsupervised machine learning techniques
• Automate predictor selection, cross validation, and hyperparameter tuning
• Graphically build and debug a deep neural network
• Access existing deep learning models from TensorFlow, Caffe and PyTorch in MATLAB
• Scaling and increasing performance with multiple processors/GPUs, clusters, and the Cloud
• Automated GPU code generation for real-time production applications
• Deployment of models to big data platforms such as Hadoop and Spark

Who Should Attend
• Data Scientists, Engineers, and Architects
• Business Intelligence Analysts
• Quantitative Analysts
• Financial Engineers
• Risk and Portfolio Managers
• Analytical Researchers
• Economists