Uploaded August 2026 | Updated November 2026, 3 days ago
Book: Fundamentals of Active Inference
Principles, Algorithms, and Applications of the Free Energy Principle for Engineers, by Sanjeev V. Namjoshi https://mitpress.mit.edu/9780262050951/fundamentals-of-active-inference/
Register: http://textbook-group.activeinference.institute/
Code: github.com/ActiveInferenceInstitute/fundamentals
Playlist of Textbook Group: youtube.com/playlist?list=PLNm0u2n1IwdrpoTg-9l37YIBiDrrOzTw-
Direct link to group document: coda.io/d/_dZHp_UGhVZE/Meetings_su3ToqMp
------
Active Inference Institute information:
Website: https://activeinference.institute/
X: https://x.com/InferenceActive
Discord: https://discord.activeinference.institute/
Donate: https://donate.activeinference.institute/
YouTube: youtube.com/c/ActiveInference
Active Inference Livestreams: https://video.activeinference.institute/
------
CHAPTERS
0:00 Welcome; last session on Chapter 8
1:30 Formalizing the hierarchical model
3:00 Probability densities across layers
4:30 8.3 Inference in hierarchical models
6:00 Precision-weighted prediction errors across layers
7:30 Bookkeeping: prediction errors by layer (page 194)
9:00 State neurons and error neurons at each layer
10:30 Repeated units across the hierarchy
12:00 Assumption: generalized coordinates for states only
15:00 Notes on beliefs about precisions
16:30 The Laplace approximation: tracking means
18:00 Free energy minimization in the multivariate case
19:30 Embedding orders (gamma)
22:30 Recap of the univariate case
25:30 Erratum: damping terms in Eq. 8.49
27:00 8.4 Generalized filtering for hierarchical models
30:00 The culmination of Chapters 6 to 8
31:30 Q&A: embedding orders vs. time steps
34:30 Figure: state units and error units
39:00 Hierarchical message passing
40:30 Excitatory and inhibitory connections
42:00 Links to Eqs. 5.44
43:30 Layers, hidden states and log precisions in the figure
46:30 Many-layer models
51:00 Connection strengths as synaptic efficacy (Figure 8.10)
52:30 8.5 Action: the forward model
55:30 Erratum: arrow directions in the forward pass
57:00 From sensory evidence to action
58:30 8.6 Summary and reading recommendations
1:00:00 Discussion: discrete and continuous models together
1:03:00 Discussion: models for clinical and experimental work
1:06:00 Algorithm 12 without action or generalized coordinates
1:07:30 Structure learning
1:09:00 Discussion: embedding orders vs. future time steps
1:16:30 Running the Chapter 8 code examples
Book: Fundamentals of Active Inference
Principles, Algorithms, and Applications of the Free Energy Principle for Engineers, by Sanjeev V. Namjoshi https://mitpress.mit.edu/9780262050951/fundamentals-of-active-inference/
Register: http://textbook-group.activeinference.institute/
Code: github.com/ActiveInferenceInstitute/fundamentals
Playlist of Textbook Group: youtube.com/playlist?list=PLNm0u2n1IwdrpoTg-9l37YIBiDrrOzTw-
Direct link to group document: coda.io/d/_dZHp_UGhVZE/Meetings_su3ToqMp
------
Active Inference Institute information:
Website: https://activeinference.institute/
X: https://x.com/InferenceActive
Discord: https://discord.activeinference.institute/
Donate: https://donate.activeinference.institute/
YouTube: youtube.com/c/ActiveInference
Active Inference Livestreams: https://video.activeinference.institute/
------
CHAPTERS
0:00 Welcome; last session on Chapter 8
1:30 Formalizing the hierarchical model
3:00 Probability densities across layers
4:30 8.3 Inference in hierarchical models
6:00 Precision-weighted prediction errors across layers
7:30 Bookkeeping: prediction errors by layer (page 194)
9:00 State neurons and error neurons at each layer
10:30 Repeated units across the hierarchy
12:00 Assumption: generalized coordinates for states only
15:00 Notes on beliefs about precisions
16:30 The Laplace approximation: tracking means
18:00 Free energy minimization in the multivariate case
19:30 Embedding orders (gamma)
22:30 Recap of the univariate case
25:30 Erratum: damping terms in Eq. 8.49
27:00 8.4 Generalized filtering for hierarchical models
30:00 The culmination of Chapters 6 to 8
31:30 Q&A: embedding orders vs. time steps
34:30 Figure: state units and error units
39:00 Hierarchical message passing
40:30 Excitatory and inhibitory connections
42:00 Links to Eqs. 5.44
43:30 Layers, hidden states and log precisions in the figure
46:30 Many-layer models
51:00 Connection strengths as synaptic efficacy (Figure 8.10)
52:30 8.5 Action: the forward model
55:30 Erratum: arrow directions in the forward pass
57:00 From sensory evidence to action
58:30 8.6 Summary and reading recommendations
1:00:00 Discussion: discrete and continuous models together
1:03:00 Discussion: models for clinical and experimental work
1:06:00 Algorithm 12 without action or generalized coordinates
1:07:30 Structure learning
1:09:00 Discussion: embedding orders vs. future time steps
1:16:30 Running the Chapter 8 code examples










