Realign schedule: W7 time-series self-study, W8 inverse problems

Tue 26.05 lecture cancelled (Pfingstdienstag) -> Week 7 time-series
becomes self-study with in-class Thu exercise. Thu 04.06 exercise
cancelled (Fronleichnam) -> Week 8 is inverse problems with a
self-study exercise. Generalization unit dropped (duplicate); Weeks
9-13 shifted up one; Week 14 is now a buffer/review/mini-project slot.
Slide links updated to the correct deck folders.

Author: PhilippPelz

index.qmd

@@ -314,14 +314,16 @@ Fine-tune a pretrained model; compare against training from scratch.
 
 ---
 
-### Unit III — Learning from Processing Data (Weeks 7–9)
+### Unit III — Learning from Processing Data (Weeks 7–8)
 
 #### Week 7 – Time-series and process monitoring
 
-*Lecture: Tuesday, 26.05.2026, 14:15-15:45 (cancelled - Pfingstdienstag) | Exercise: Thursday, 28.05.2026, 16:15-17:45*
+*Lecture: Tuesday, 26.05.2026, 14:15-15:45 (self-study — Pfingstdienstag public holiday) | Exercise: Thursday, 28.05.2026, 16:15-17:45 (in class)*
 
 **Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit07_time_series/01_intro.html)
 
+> **Self-study lecture:** the Tuesday slot is cancelled (Pfingstdienstag). Work through the slide deck independently; the Thursday exercise runs in class and consolidates the material.
+
 - Processing signals:
   temperature cycles, AM melt pool signals, SPS, rolling.
 - Regression and sequence models as surrogates.
@@ -335,41 +337,18 @@ Fine-tune a pretrained model; compare against training from scratch.
 - Case studies: additive manufacturing, process stability
 - Real-time anomaly detection from processing history
 
-
-**Exercise:**  
+**Exercise:**
 Predict a process outcome from time-series data using regression or simple RNNs.
 
 ---
 
-#### Week 8 – Generalization, robustness, and process windows
-
-*Lecture: Tuesday, 02.06.2026, 14:15-15:45 | Exercise: Thursday, 04.06.2026, 16:15-17:45 (cancelled - Fronleichnam)*
-
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit08_generalization_robustness/01_intro.html)
-
-- Sensitivity to noise and parameter drift.
-- Overfitting in process–property models.
-- Robustness as a design criterion.
-
-**Summary:**
-
-- Shift from raw performance to **model reliability**
-- Bias–variance tradeoff and generalization to factory-floor data
-- Robust validation: K-fold and stratified cross-validation on small datasets
-- **Process robustness** via sensitivity analysis
-- **Process windows**: parameter regions insensitive to industrial noise
-
+#### Week 8 – Inverse problems and process maps
 
-**Exercise:**  
-Analyze model robustness under perturbed process conditions.
-
----
+*Lecture: Tuesday, 02.06.2026, 14:15-15:45 (in class) | Exercise: Thursday, 04.06.2026, 16:15-17:45 (self-study — Fronleichnam public holiday)*
 
-#### Week 9 – Inverse problems and process maps
+**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit08_inverse_problems/09_inverse_problems.html)
 
-*Lecture: Tuesday, 09.06.2026, 14:15-15:45 | Exercise: Thursday, 11.06.2026, 16:15-17:45*
-
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit09_inverse_problems/09_inverse_problems.html)
+> **Self-study exercise:** the Thursday slot is cancelled (Fronleichnam). The exercise is provided for independent work; a solution is released afterwards.
 
 - Process → structure inverse problems.
 - ML-guided process maps (e.g. AM laser power vs scan speed).
@@ -382,19 +361,20 @@ Analyze model robustness under perturbed process conditions.
 - **Physics-informed learning**: physical transformations and constraints
 - **Process maps** and **process corridors** for safe operating regions
 
-
-**Exercise:**  
+**Exercise:**
 Construct a simple ML-based process map; compare constrained vs unconstrained models.
 
 ---
 
-### Unit IV — Uncertainty, Surrogates, and Automation (Weeks 10–12)
+### Unit IV — Characterization, Automation, and Uncertainty (Weeks 9–11)
 
-#### Week 10 – ML for characterization signals
+#### Week 9 – ML for characterization signals
 
-*Lecture: Tuesday, 16.06.2026, 14:15-15:45 | Exercise: Thursday, 18.06.2026, 16:15-17:45*
+*Lecture: Tuesday, 09.06.2026, 14:15-15:45 | Exercise: Thursday, 11.06.2026, 16:15-17:45*
+
+**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit09_characterization_signals/10_characterization_signals.html)
 
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit10_characterization_signals/10_characterization_signals.html)
+Companion deck (within Week 9): **Transformers for materials** — [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit09b_transformers_for_materials/transformers_for_materials.html)
 
 - Spectral data: XRD, EELS, EDS.
 - Denoising, peak finding, dimensionality reduction.
@@ -407,17 +387,16 @@ Construct a simple ML-based process map; compare constrained vs unconstrained mo
 - **Autoencoders**: compressing spectra into a low-dimensional latent space
 - Denoising and feature extraction at high throughput without losing physics
 
-
-**Exercise:**  
+**Exercise:**
 Apply PCA/NMF to spectral datasets; interpret components physically.
 
 ---
 
-#### Week 11 – Automation in microscopy and characterization
+#### Week 10 – Automation in microscopy and characterization
 
-*Lecture: Tuesday, 23.06.2026, 14:15-15:45 | Exercise: Thursday, 25.06.2026, 16:15-17:45*
+*Lecture: Tuesday, 16.06.2026, 14:15-15:45 | Exercise: Thursday, 18.06.2026, 16:15-17:45*
 
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit11_automation/11_automation.html)
+**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit10_automation/11_automation.html)
 
 - Autofocus, drift correction, parameter selection.
 - ML as a control component, not just a predictor.
@@ -434,54 +413,62 @@ Implement a simple ML-assisted autofocus or defect detector.
 
 ---
 
-#### Week 12 – Uncertainty-aware regression & Gaussian Processes
+#### Week 11 – Uncertainty-aware regression & Gaussian Processes
 
-*Lecture: Tuesday, 30.06.2026, 14:15-15:45 | Exercise: Thursday, 02.07.2026, 16:15-17:45*
+*Lecture: Tuesday, 23.06.2026, 14:15-15:45 | Exercise: Thursday, 25.06.2026, 16:15-17:45*
 
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit12_uncertainty_gp/12_uncertainty_gp.html)
+**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit11_uncertainty_gp/12_uncertainty_gp.html)
 
 - Aleatoric vs epistemic uncertainty in experiments.
 - Gaussian Processes as uncertainty-aware surrogates.
 - Exploration vs exploitation in experimental design.
 - Connection to materials acceleration platforms.
 
-**Exercise:**  
+**Exercise:**
 Compare GP regression and NN ensembles for a process-parameter problem.
 
 ---
 
-### Unit V — Physics, Trust, and Synthesis (Weeks 13–14)
+### Unit V — Physics, Trust, and Synthesis (Weeks 12–13)
 
-#### Week 13 – Physics-informed and constrained ML
+#### Week 12 – Physics-informed and constrained ML
 
-*Lecture: Tuesday, 07.07.2026, 14:15-15:45 | Exercise: Thursday, 09.07.2026, 16:15-17:45*
+*Lecture: Tuesday, 30.06.2026, 14:15-15:45 | Exercise: Thursday, 02.07.2026, 16:15-17:45*
 
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit13_pinns/13_pinns.html)
+**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit12_pinns/13_pinns.html)
 
 - Embedding physical constraints into ML models.
 - Penalty terms, soft constraints, hybrid approaches.
 - Failure modes of unconstrained models.
 
-**Exercise:**  
+**Exercise:**
 Train a constrained model for a processing or characterization task.
 
 ---
 
-#### Week 14 – Integration, limits, and reflection
+#### Week 13 – Integration, limits, and reflection
 
-*Lecture: Tuesday, 14.07.2026, 14:15-15:45 | Exercise: Thursday, 16.07.2026, 16:15-17:45*
+*Lecture: Tuesday, 07.07.2026, 14:15-15:45 | Exercise: Thursday, 09.07.2026, 16:15-17:45*
 
-**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit14_reflection/14_reflection.html)
+**Slides:** [Open](https://pelzlab.science/public_presentations/ml_for_characterization_and_processing/unit13_reflection/14_reflection.html)
 
 - Explainability for experimental ML (CAMs, SHAP).
 - Why ML fails in real labs.
 - Where ML genuinely changes materials processing.
 
-**Exercise:**  
+**Exercise:**
 Mini-project presentations and critical discussion.
 
 ---
 
+#### Week 14 – Buffer, review, and mini-project work
+
+*Tuesday, 14.07.2026, 14:15-15:45 | Thursday, 16.07.2026, 16:15-17:45*
+
+No new material. Reserved as a buffer to absorb schedule slippage from the Week 7 / Week 8 public-holiday self-study sessions, for review of difficult topics on request, and for mini-project consultation and presentations.
+
+---
+
 ## Learning Outcomes
 
 Students completing this course will be able to:

week10_summary.md

@@ -1,50 +1,47 @@
-# Week 10 Summary: ML for characterization signals
+# Week 10 Summary: Automation in microscopy and characterization
 
 ## Cross-Book Summary
 
-### 1. Clustering Spectral Data
-- **K-Means:** Groups similar spectra (XRD/EDS) to identify distinct phases.
-- **Mini-Batch K-Means:** Speeds up high-throughput characterization.
-- **t-SNE:** Projects high-dimensional spectra to 2D to reveal outliers/relationships.
+### 1. Multi-Modal Data Fusion
+- **Beyond Single Sensors:** Fuse images (SEM), chemistry (EDS), and orientations (EBSD) for a complete physical picture.
+- **Bayesian Sensor Fusion:** Combines uncertain measurements using precision-weighted posteriors.
+- **Latent Fusion:** Autoencoders/PCA find shared embeddings to combine diverse data types.
 
-### 2. Autoencoders for Signal Processing
-- **Latent Representations:** Compresses spectra to essential physical information.
-- **Denoising:** Reconstructs clean signals from noisy inputs without blurring.
-- **Non-linear Compression:** Outperforms PCA for complex spectral libraries.
+### 2. Reinforcement Learning for Control
+- **Autonomous Agent:** Learns to interact with environments (e.g., microscopes) to maximize rewards.
+- **RL Loop:** State (image), Action (adjust focus), Reward (sharpness/SNR).
+- **Policy Gradients:** Train NNs for optimal scientific decision-making.
 
-### 3. Scientific Integrity in ML
-- **Peak Preservation:** ML must assist, not invent or smooth away real physics.
+### 3. Computer Vision in the Lab
+- **Automated Workflows:** CNNs for real-time ROI detection, autofocus, and pattern classification.
 
 ## 90-Minute Lecture Strategy
 
-### Part 1: High-Dimensional Signals
-- Digital footprint: XRD, EDS, EELS, Raman.
-- Manual vs. automated peak-picking.
-- Vector spectrum representation.
+### Part 1: Toward the Self-Driving Lab
+- The automation stack.
+- Autonomous Characterization: Scan, Analyze, Decide, Repeat.
 
-### Part 2: Clustering Structure
-- K-Means algorithm.
-- Elbow Method for phase counting.
-- Ternary alloy mapping.
+### Part 2: ML-Assisted Instrument Tuning
+- Autofocus and Beam Alignment.
+- Real-time feedback loops.
 
-### Part 3: Visualizing the Unseen
-- t-SNE Stochastic Proximity.
-- Hidden relationships.
-- t-SNE distance pitfalls.
+### Part 3: Fusing Multi-Modal Data
+- Bayesian Fusion for sensor noise.
+- Multi-head NNs.
+- Combining XRD and EDS.
 
-### Part 4: Autoencoders & Denoising
-- Encoder-Bottleneck-Decoder.
-- Denoising characterization signals.
-- Bottlenecks as physical descriptors.
+### Part 4: RL for Lab Control
+- RL Framework overview.
+- Reward Functions for science.
+- Industrial glass processing control.
 
-### Part 5: Data to Discovery
-- Real-time spectral analysis.
-- Physical consistency in ML.
-- Automated pipelines.
+### Part 5: The Integrated Pipeline
+- "On-the-fly" discovery.
+- Automation challenges: Latency and safety.
 
 ## Quarto Website Update (Summary)
-**Summary for ML-PC Week 10:**  
-- Processes high-dimensional Characterization Signals (XRD, EDS).
-- Employs K-Means and t-SNE for automated phase identification.
-- Uses Autoencoders for latent space compression and denoising.
-- Enhances high-throughput data analysis while preserving physics.
+**Summary for ML-PC Week 11:**  
+- Explores Autonomous Characterization and active instrument control.
+- Introduces Multi-Modal Data Fusion (Bayesian and Latent).
+- Uses Reinforcement Learning (RL) for laboratory task automation.
+- Details building integrated pipelines for "on-the-fly" scientific discovery.

week11_summary.md

@@ -1,47 +1,47 @@
-# Week 11 Summary: Automation in microscopy and characterization
+# Week 11 Summary: Uncertainty-aware regression & Gaussian Processes
 
 ## Cross-Book Summary
 
-### 1. Multi-Modal Data Fusion
-- **Beyond Single Sensors:** Fuse images (SEM), chemistry (EDS), and orientations (EBSD) for a complete physical picture.
-- **Bayesian Sensor Fusion:** Combines uncertain measurements using precision-weighted posteriors.
-- **Latent Fusion:** Autoencoders/PCA find shared embeddings to combine diverse data types.
+### 1. Knowing what you don't know
+- **Aleatoric vs. Epistemic:** Inherent physical noise vs. model ignorance.
+- **Overconfidence Danger:** Point estimates fail safely in unknown regimes; uncertainty metrics are crucial.
 
-### 2. Reinforcement Learning for Control
-- **Autonomous Agent:** Learns to interact with environments (e.g., microscopes) to maximize rewards.
-- **RL Loop:** State (image), Action (adjust focus), Reward (sharpness/SNR).
-- **Policy Gradients:** Train NNs for optimal scientific decision-making.
+### 2. Gaussian Processes (GPs)
+- **Distribution over Functions:** GP yields posterior mean and variance (uncertainty).
+- **Kernels as Physical Priors:** Encodes assumptions about data smoothness/scale.
+- **Non-Parametric Nature:** Scales with data size, ideal for small, high-quality materials datasets.
 
-### 3. Computer Vision in the Lab
-- **Automated Workflows:** CNNs for real-time ROI detection, autofocus, and pattern classification.
+### 3. GP-Based Process Maps
+- **Confidence Ribbons:** Visualize reliability to guide further experiments.
+- **Kriging:** Interpolates materials property surfaces using GP regression.
 
 ## 90-Minute Lecture Strategy
 
-### Part 1: Toward the Self-Driving Lab
-- The automation stack.
-- Autonomous Characterization: Scan, Analyze, Decide, Repeat.
+### Part 1: Uncertainty in Science
+- Risk management in materials processing.
+- Visualizing distributions and error bars.
 
-### Part 2: ML-Assisted Instrument Tuning
-- Autofocus and Beam Alignment.
-- Real-time feedback loops.
+### Part 2: GP Fundamentals
+- Function vs. Parameter space.
+- Kernels and "Similarity".
+- Conditional Gaussians and Variance.
 
-### Part 3: Fusing Multi-Modal Data
-- Bayesian Fusion for sensor noise.
-- Multi-head NNs.
-- Combining XRD and EDS.
+### Part 3: GP Case Studies
+- Predicting tensile strength across parameters.
+- GP for Experimental Design.
+- Multi-Task GPs.
 
-### Part 4: RL for Lab Control
-- RL Framework overview.
-- Reward Functions for science.
-- Industrial glass processing control.
+### Part 4: Advanced Probabilistic ML
+- Mixture Density Networks (MDNs).
+- Dropout as Bayesian approximation.
 
-### Part 5: The Integrated Pipeline
-- "On-the-fly" discovery.
-- Automation challenges: Latency and safety.
+### Part 5: Decision Making
+- Safe process windows via confidence intervals.
+- Building trustworthy models.
 
 ## Quarto Website Update (Summary)
-**Summary for ML-PC Week 11:**  
-- Explores Autonomous Characterization and active instrument control.
-- Introduces Multi-Modal Data Fusion (Bayesian and Latent).
-- Uses Reinforcement Learning (RL) for laboratory task automation.
-- Details building integrated pipelines for "on-the-fly" scientific discovery.
+**Summary for ML-PC Week 12:**  
+- Introduces Probabilistic Machine Learning for uncertainty quantification.
+- Differentiates aleatoric (noise) from epistemic (ignorance) uncertainty.
+- Uses Gaussian Processes (GPs) for uncertainty-aware regression.
+- Applies confidence intervals to map robust process windows.