This lecture will cover the mathematical foundations of deep learning; optimization algorithms such as stochastic gradient descent, RMSprop, and ADAM; computational methods for scaling deep learning to massive datasets and models; and deep reinforcement learning. Several applications of deep learning for modeling financial data will be presented, including: loan delinquency/prepayment modeling, high-frequency order book data, and solving high-dimensional options. The potential of deep learning to contribute to the financial sector across a number of key areas (e.g., lending, hedging, pricing, market making) will be discussed.