Uploaded October 2026 | Updated November 2026, 3 hours ago
How should a limited monitoring budget be distributed across California?
Explore Active Inference through a statewide example spanning fire, seismic activity, and drought. Animated graphical models connect hidden conditions, observations, belief updates, and sensing decisions to a practical question: what should we observe next?
This walkthrough follows the California thin orchestrator in GEO-INFER’s PLACE module, connecting geospatial execution, explicit generative models, and mathematical reference across three Active Inference Institute repositories.
WHAT YOU’LL LEARN
• How Active Inference connects beliefs, predictions, observations, and actions.
• How sensor likelihoods distinguish missing, negative, and positive reports.
• How expected free energy balances information gain against preference cost.
• How passive, coarse, and detailed sensing modes compete for a shared budget.
• How geography, clipped-cell areas, county shares, and spatial coupling affect interpretation.
• How active, quota-active, and rotating strategies compare across monitoring rounds.
• How thin orchestration, model interchange, retained artifacts, and validation support reproducible research.
The example uses historical environmental indicators and simulated sensor reports. Their roles remain explicit throughout, alongside the limits of approximate spatial inference, numerical comparisons, and formal mathematical claims.
CHAPTERS
00:00 The California question and Active Inference overview
05:05 Sources, time, and geography
06:57 The local generative model
09:02 Policies and a shared monitoring budget
10:57 The complete executable workflow
13:06 Reading the measured comparison
15:15 Coverage, uncertainty, and formal scope
17:32 Reproduce, inspect, and extend
CODE AND DOCUMENTATION
GEO_INFER
Geospatial inference, PLACE workflows, and the California implementation:
github.com/ActiveInferenceInstitute/GEO_INFER
GEO-INFER-PLACE module:
github.com/ActiveInferenceInstitute/GEO_INFER/tree/main/GEO-INFER-PLACE
California methods and source guide:
github.com/ActiveInferenceInstitute/GEO_INFER/blob/main/GEO-INFER-INTRA/docs/examples/california_statewide_monitoring.md
Generalized_Notation_Notation
Explicit state-space models, model checking, and interchange:
github.com/ActiveInferenceInstitute/Generalized_Notation_Notation
FEP_Formal
Formal mathematical foundations, theorem statements, and explicit assumptions:
github.com/ActiveInferenceInstitute/fep_formal
ACTIVE INFERENCE INSTITUTE
Explore education, research, open-source projects, and ways to participate:
https://activeinference.institute/
#ActiveInference #Geospatial #GEOINFER
How should a limited monitoring budget be distributed across California?
Explore Active Inference through a statewide example spanning fire, seismic activity, and drought. Animated graphical models connect hidden conditions, observations, belief updates, and sensing decisions to a practical question: what should we observe next?
This walkthrough follows the California thin orchestrator in GEO-INFER’s PLACE module, connecting geospatial execution, explicit generative models, and mathematical reference across three Active Inference Institute repositories.
WHAT YOU’LL LEARN
• How Active Inference connects beliefs, predictions, observations, and actions.
• How sensor likelihoods distinguish missing, negative, and positive reports.
• How expected free energy balances information gain against preference cost.
• How passive, coarse, and detailed sensing modes compete for a shared budget.
• How geography, clipped-cell areas, county shares, and spatial coupling affect interpretation.
• How active, quota-active, and rotating strategies compare across monitoring rounds.
• How thin orchestration, model interchange, retained artifacts, and validation support reproducible research.
The example uses historical environmental indicators and simulated sensor reports. Their roles remain explicit throughout, alongside the limits of approximate spatial inference, numerical comparisons, and formal mathematical claims.
CHAPTERS
00:00 The California question and Active Inference overview
05:05 Sources, time, and geography
06:57 The local generative model
09:02 Policies and a shared monitoring budget
10:57 The complete executable workflow
13:06 Reading the measured comparison
15:15 Coverage, uncertainty, and formal scope
17:32 Reproduce, inspect, and extend
CODE AND DOCUMENTATION
GEO_INFER
Geospatial inference, PLACE workflows, and the California implementation:
github.com/ActiveInferenceInstitute/GEO_INFER
GEO-INFER-PLACE module:
github.com/ActiveInferenceInstitute/GEO_INFER/tree/main/GEO-INFER-PLACE
California methods and source guide:
github.com/ActiveInferenceInstitute/GEO_INFER/blob/main/GEO-INFER-INTRA/docs/examples/california_statewide_monitoring.md
Generalized_Notation_Notation
Explicit state-space models, model checking, and interchange:
github.com/ActiveInferenceInstitute/Generalized_Notation_Notation
FEP_Formal
Formal mathematical foundations, theorem statements, and explicit assumptions:
github.com/ActiveInferenceInstitute/fep_formal
ACTIVE INFERENCE INSTITUTE
Explore education, research, open-source projects, and ways to participate:
https://activeinference.institute/
#ActiveInference #Geospatial #GEOINFER










![ActInf GuestStream 115.1 ~ Energy-Based Transformers and the Future of Scaling
Energy-Based Transformers are Scalable Learners and Thinkers
Alexi Gladstone, Ganesh Nanduru, Md Mofijul Islam, Peixuan Han, Hyeonjeong Ha, Aman Chadha, Yilun Du, Heng Ji, Jundong Li, Tariq Iqbal
https://arxiv.org/abs/2507.02092
Inference-time computation techniques, analogous to human System 2 Thinking, have recently become popular for improving model performances. However, most existing approaches suffer from several limitations: they are modality-specific (e.g., working only in text), problem-specific (e.g., verifiable domains like math and coding), or require additional supervision/training on top of unsupervised pretraining (e.g., verifiers or verifiable rewards). In this paper, we ask the question Is it possible to generalize these System 2 Thinking approaches, and develop models that learn to think solely from unsupervised learning? Interestingly, we find the answer is yes, by learning to explicitly verify the compatibility between inputs and candidate-predictions, and then re-framing prediction problems as optimization with respect to this verifier. Specifically, we train Energy-Based Transformers (EBTs) a new class of Energy-Based Models (EBMs) to assign an energy value to every input and candidate-prediction pair, enabling predictions through gradient descent-based energy minimization until convergence. Across both discrete (text) and continuous (visual) modalities, we find EBTs scale faster than the dominant Transformer++ approach during training, achieving an up to 35% higher scaling rate with respect to data, batch size, parameters, FLOPs, and depth. During inference, EBTs improve performance with System 2 Thinking by 29% more than the Transformer++ on language tasks, and EBTs outperform Diffusion Transformers on image denoising while using fewer forward passes. Further, we find that EBTs achieve better results than existing models on most downstream tasks given the same or worse pretraining performance, suggesting that EBTs generalize better than existing approaches. Consequently, EBTs are a promising new paradigm for scaling both the learning and thinking capabilities of models.
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2507.02092 [cs.LG]
(or arXiv:2507.02092v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2507.02092
Active Inference Institute information:
Website: https://www.activeinference.institute/
Activities: https://activities.activeinference.institute/
Discord: https://discord.activeinference.institute/
Donate: http://donate.activeinference.institute/
YouTube: https://www.youtube.com/c/ActiveInference/
X: https://x.com/InferenceActive
Active Inference Livestreams: https://video.activeinference.institute/ ActInf GuestStream 115.1 ~ Energy-Based Transformers and the Future of Scaling](https://i.ytimg.com/vi/85ob-jrG7xQ/mqdefault.jpg)