Deep Learning — Goodfellow, Bengio & Courville

Reading

Where I am: page 87 of 800 — Chapter 4, Numerical Computation.

Notes so far

• Ch. 2 — Linear Algebra: eigendecomposition and SVD as the core lenses for understanding data transformations.
• Ch. 3 — Probability: why ML prefers probability distributions over hard rules; information theory basics.
• Ch. 4 — Numerical Computation: currently reading — overflow/underflow, conditioning, and gradient-based optimization.

To update my progress: change currentPage in the META block at the top of this file, rebuild, and push.