Join us for an advanced deep learning workshop hosted by the NVIDIA Deep Learning Institute and the Harvard Big Data Club. This all-day event will feature 2 hands-on labs, one in the morning and another following a lunch break.

*We will be raffling an NVIDIA Shield TV!

LABS

I. Modeling Time Series Data with Recurrent Neural Networks in Keras

Prerequisites: Some experience training CNNs
Frameworks: Theano

Recurrent Neural Networks (RNNs) allow models to classify or forecast time-series data, like natural language, markets, and in the case of this lab, a patient’s health over time. You will:

• Create training and testing data sets using electronic health records in HDF5 (hierarchical data format version five)
• Prepare data sets for use with recurrent neural networks (RNNs), which allows modeling of very complex data sequences
• Construct a long-short term memory model (LSTM), a specific RNN architecture, using the Keras library running on top of Theano to evaluate model performance against baseline data
• Upon completion, you will be able to model time-series data using Recurrent Neural Networks

II. Algorithmic Trading Using Deep Autoencoder-Based Statistical Arbitrage

Prerequisites: Fundamentals of Deep Learning with Computer Vision or similar experience
Framework: Keras with TensorFlow backend

Learn to reconstruct security return data and use anomalous deviations (spreads) in their errors as signals for creating long/short positions. Use Keras with Tensorflow backend and Python to:

• Structure and train a deep autoencoder
• Use the autoencoder as an anomaly detector to create an arbitrage strategy
• Perform a hyperparameter search over the autoencoder
• Upon completion, you’ll be able to implement a deep learning strategy for algorithmic training and calculate its P&L

AGENDA

9:30am-10:00am: Check-in, set-up, and breakfast
10:00am-10:30am: Introductions
10:30am-12:30pm: Lab 1 - Modelling Time Series
12:30-1:00pm: Lunch Break and Networking
1:00pm-3:00pm: Lab 2 – Algorithmic Trading
3:00pm-3:30pm: Wrap-up and raffle

LOCATION

Location will be emailed out to all registrants the week of the event.

FAQ

What are the prerequisites?
This is a workshop at an advanced level and participants are expected to have strong technical experience.

Can I register for just one session?
Participants are expected to attend both workshops.

What do I need to bring?
Your own fully charged laptop and a charger (just in case).

Is this open to the public?
Preference will be given to Harvard affiliates.

ATTENDEE SET-UP INSTRUCTIONS
You must bring your own laptop to these labs.

Before getting started, follow these steps to set up for the session:

1. Create a Qwiklabs account at https://nvlabs.qwiklab.com/
2. Ensure Qwiklabs runs smoothly on your laptop by going to http://websocketstest.com/
3. Make sure that WebSockets work for you by checking:
- Under ENVIRONMENT, WebSockets is checked yes.
- Under WEBSOCKETS (PORT 80), Data Receive, Send, and Echo Test are checked yes.
4. If there are issues with WebSockets, try updating your browser. Best browsers for Qwiklabs are Chrome, FireFox, and Safari. The labs will run in Internet Explorer, but you will not have an optimal experience.
5. You’re ready to get started.