Human and
Machine Learning

Learn how the mind works by building it — from Bayesian inference and causal reasoning to reinforcement learning and neural networks, with hands-on probabilistic programming.

Spring 2026
/
Fridays, Apr 17 – Jul 17
/
Prof. Joseph Austerweil

All interested graduate students at Chiba Tech are encouraged to enroll

Permission Enrollment is by instructor permission only. Please reach out to Prof. Austerweil via email if you are interested.
Language This course is taught entirely in English. English proficiency is required.

joseph.austerweil@chibatech.ac.jp

Course Highlights

A graduate seminar bridging cognitive science, machine learning, and probabilistic programming.

Bayesian Foundations

Build intuition for probability from counting to conjugate models, hierarchical Bayes, and Bayesian nonparametrics.

Hands-on GenJAX

Write probabilistic programs from Week 2 onward. Model, condition, and run inference using GenJAX on JAX — no toy examples.

Cognitive Models That Explain

Causal reasoning, generalization, reinforcement learning, and social cognition — framed as computational-level theories of the mind.

Contemporary ML Connections

Bridge to transformers, scaling laws, RLHF, and AI alignment — see where classical models meet modern deep learning.

Living Textbook

A free, open-source textbook on probability and probabilistic computing that grows alongside the course.

Discussion-Driven

Weekly reflections, in-class exercises, and collaborative problem-solving — not just lectures.

Free Textbook

A Narrative Introduction to Probability covers discrete & continuous probability, Bayesian learning, mixture models, and probabilistic programming with GenJAX — all with interactive notebooks.

Read the Textbook GitHub Repo

Weekly Schedule

12 sessions, Fridays. Topics build cumulatively; GenJAX integration is woven throughout.

WkDateTopic
1 Apr 17 Introduction & Basic Bayes PDF Bk Hungry Bk Probability As Counting Bk Getting Started Bk Python Basics
2 Apr 24 Levels of Analysis & Bayes cont'd PDF Bk Conditional Probability Bk Bayes Rule Bk Mystery Bentos Bk Continuous Bk Gaussian T2 Ch 0 T2 Ch 2
May 1No class
May 8No class (holiday)
3 May 15 Conjugate Bayes & Topic Models PDF Bk Bayesian Learning Bk Glossary Bk First Model Bk Traces
4 May 22 Generalization & Hierarchical Bayes PDF T3 Bayesian Learning T3 Mixture Models T3 Generalization T3 Hierarchical Bayes T2 Conditioning
5 May 29 Bayes Nets & Causal Bayes Nets PDF T3 Mixture Models T3 Hierarchical Bayes T2 Building Models
6 Jun 5 Markov Chains & Networks PDF T3 Markov Chains T3 Random Walks Networks T3 Memory Search
7 Jun 12 Monte Carlo Methods PDF T3 Ch 12 T3 Monte Carlo T3 Particle Filtering T3 Markov Chain Monte Carlo T3 Sampling The Mind
8 Jun 19 SDT, MDPs & Reinforcement Learning PDF Decision theory Loss → estimator MDPs & Bellman Value iteration & γ Q-learning Reward shaping Simulation-based RL & MCTS Two-step task (MB vs MF)
9 Jun 26 Inverse Reinforcement Learning PDF T3 Inverse Rl Goal Inference T3 Pomdps Belief Inference T3 Modern Rl World Models
10 Jul 3 Bayesian Nonparametrics PDF Bias-variance dilemma Ridge = a Gaussian prior Double descent Discrete BNP: DP / CRP / DPMM One object, three lenses Gaussian processes GP → NNGP / NTK It all comes home
11 Jul 10 Deep Neural Networks PDF Bk Vectors And Spaces Bk From Rules To Weights Bk Neural Net Fundamentals Bk Transformers Attention Bk Llms In Context Learning Bk World Models Imagination
12 Jul 17 Ethics & Adversarial ML PDF Bk Adversarial Examples Bk Fairness Formalisms Bk Bias In Data Bk Alignment Safety

Grading

  • Final project — 50%
  • Programming assignments (4) — 30%
  • Weekly discussion posts — 12.5%
  • Paper presentation — 7.5%

Full breakdown on the syllabus.

Assignments

  • Clusters (mixture models)
  • Bayesian Generalization
  • Monte Carlo Estimation
  • Reinforcement Learning

All four assignments are completed in GenJAX.

Prerequisites

  • Graduate standing or instructor consent
  • Comfort with basic probability & statistics
  • Programming experience (Python preferred)
  • Curiosity about how minds compute

Resources

  • Free online textbook with Colab notebooks
  • Weekly readings from primary literature
  • Probability cheatsheet
  • GenJAX setup guide & tutorials