This book offers a thorough introduction to probabilistic deep learning, merging the power of neural network architectures with statistical modeling principles. It covers foundational deep learning concepts such as fully connected and convolutional neural networks, gradient descent, and backpropagation, before delving into how to build robust loss functions using maximum likelihood and other probabilistic approaches. Readers will learn to develop advanced models for classification, regression, and complex data distributions, exploring practical applications with tools like TensorFlow Probability and techniques such as normalizing flows, ultimately gaining both the theoretical understanding and practical skills to create high-performing and interpretable deep learning solutions.
This page is free — overview and chapter list only.
The complete book body is sold separately.
Full book: $2.99 USDC via x402 ·
Per chapter: $0.25 USDC
Buy / open complete book (HTML)
· Complete book Markdown (.md)
Agents: start with free /library/discovery.json, sample free teaser chapters,
then pay for individual chapters or the complete book URL above.
Append .md to any content URL for Markdown with YAML front matter.
Chapters
- Part 1Basics Of Deep Learning1Page 20 0 (Free teaser)
- 311 Fitting A Linear Regression Model Based On A Loss Functi ($0.25)
- Closed Form Solution For The Optimal Parameter Estimates In ($0.25)
- 321 Loss With One Free Model Parameter ($0.25)
- The Update Rule In Gradient Descent ($0.25)
- 322 Loss With Two Free Model Parameters ($0.25)
- 331 Mini Batch Gradient Descent ($0.25)
- 332 Using Sgd Variants To Speed Up The Learning ($0.25)
- 333 Automatic Differentiation ($0.25)
- 341 Static Graph Frameworks ($0.25)
- 342 Dynamic Graph Frameworks ($0.25)
- This Chapter Covers ($0.25)
- 421 Binary Classification Problem ($0.25)
- Maxlike Approach For The Classification Loss Using A Paramet ($0.25)
- Loss Function For Classification Of Two Classes With A Singl ($0.25)
- What Entropy Means In Statistics And Information Theory ($0.25)
- 431 Using A Nn Without Hidden Layers And One Output Neuron F ($0.25)
- Recap On Normal Distributions ($0.25)
- Maxlike Based Derivation Of The Mse Loss In Linear Regressio ($0.25)
- 531 Fitting And Evaluating A Linear Regression Model With Co ($0.25)
- 532 Fitting And Evaluating A Linear Regression Model With A ($0.25)
- 541 The Poisson Distribution For Count Data ($0.25)
- Definition Of The Custom Parameterized Distributionninputs T ($0.25)
- 612 Making Sense Of Discretized Logistic Mixture ($0.25)
- 632 The Change Of Variable Technique For Probabilities ($0.25)
- 633 Fitting An Nf To Data ($0.25)
- 634 Going Deeper By Chaining Flows ($0.25)
- 636 Using Networks To Control Flows ($0.25)
- 721 Bayesian Model The Hackers Way ($0.25)
- 731 Training And Prediction With A Bayesian Model ($0.25)
- Some Fun Facts About The Choice Of Priors ($0.25)
- 733 Revisiting The Bayesian Linear Regression Model ($0.25)
- 821 Looking Under The Hood Of Vi ($0.25)
- Derivation Of The Optimization Equation ($0.25)
- 822 Applying Vi To The Toy Problem ($0.25)
- 841 Classical Dropout Used During Training ($0.25)
- 842 Mc Dropout Used During Train And Test Times ($0.25)
- 851 Regression Case Study On Extrapolation ($0.25)
- 852 Classification Case Study With Novel Classes ($0.25)