Discussion on deep autoregressive models and transformer architectures for time-series applications.

Overview of VAEs and their application to time-series data.

Overview of discriminative AI methods including CART decision trees and neural networks for trading data.

Hands-on exercise to build, fine-tune, and apply the Lag-Llama transformer to time-series data.

Hands-on exercise building a lightweight transformer for time-series prediction.

Hands-on exercise building an RNN to predict SPX returns and evaluating performance metrics.

Hands-on feature engineering with TA-LIB and building a gradient-boosted tree to predict SPX returns.

Hands-on exercise applying Gaussian Mixture Models for regime detection in time series.

Hands-on exercise applying Hidden Markov Models for regime detection in time series.

Hands-on exercise applying TimeVAE to predict SPX returns.

Comparison of generative vs discriminative AI and semi-supervised learning concepts.

Using GenAI/LLMs to create and backtest trading strategies.

In the news, GenAI is usually associated with large language models (LLMs) or with image generation tools, essentially, machines that can learn from text or images and generate text or images. But in reality, these models can learn from many different types of data. In particular, they can learn from time series of asset returns, which is perhaps the most relevant for asset managers. In this talk, and in our accompanying book (Generative AI for Trading and Asset Management), we highlight both the practical applications and the fundamental principles of GenAI, with a special focus on how these technologies apply to trading and asset management.