cohort 1, chapters 1-3 (#3)

* cohort 1, chapters 1-3 * made slides * edited "club intro" to link to our textbook * added "images" as subdirectory of "slides" * Remove extra title slide. The "1. Preliminaries" title (in the header yaml) is what shows up in the index, but it also automatically acts as the first slide. I also turned off incremental bulleted lists by default, because I find myself manually turning them off most of the time. I tried to fix the styling so you wouldn't have to insert the empty spacer columns, because those are a clear sign that something is wrong (you should be able to just put content in your qmd and let styling deal with formatting), but the revealjs style fought me. I might come back to it later. * Remove extraneous session info block. * I don't think these callout notes are displaying the way you intend. Did you preview the slides? I feel like you must have because we discussed them, but what I'm seeing looks broken. --------- Co-authored-by: Jon Harmon <[email protected]>
r4ds · Feb 1, 2025 · e0c62b2 · e0c62b2
1 parent 03115ef
commit e0c62b2
Show file tree

Hide file tree

Showing 14 changed files with 205 additions and 45 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -11,4 +11,5 @@ Depends:
     R (>= 3.1.0)
 Encoding: UTF-8
 Imports: 
+    DiagrammeR,
     rmarkdown
diff --git a/slides/00-club_intro.qmd b/slides/00-club_intro.qmd
@@ -5,7 +5,8 @@ title: Club Meetings
 
 # Welcome
 
-- Welcome to the The Rust Programming Language book club! 
+- Welcome to the Generative AI book club! 
+- "book: [Generative AI Handbook: A Roadmap for Learning Resources](https://genai-handbook.github.io/) by William Brown
 - These slides are available at [dslc.io/genai](https://dslc.io/genai).
 - We follow the [Data Science Learning Community Code of Conduct](https://dslc.io/code_of_conduct.html).
 

diff --git a/slides/01_preliminaries.qmd b/slides/01_preliminaries.qmd
@@ -3,29 +3,91 @@ engine: knitr
 title: "1. Preliminaries"
 ---
 
-# Learning objectives
+# Learning Objectives
 
-::: nonincremental
-- THESE ARE NICE TO HAVE BUT NOT ABSOLUTELY NECESSARY
+:::: {.columns}
+
+::: {.column width="45%"}
+Talk about recent history in artificial intelligence
+
+Mention mathematics prerequisites
+
+Pick a programming language
+:::
+
+::: {.column width="10%"}
+
+:::
+
+::: {.column width="45%"}
+![Generative AI](images/gen_ai_splash.png)
+
+image source: [Georgetown Journal of International Affairs](https://gjia.georgetown.edu/2023/05/24/should-the-united-states-or-the-european-union-follow-chinas-lead-and-require-watermarks-for-generative-ai/)
+
+:::
+
+::::
+
+## AI Landscape
+
+![areas for artificial intelligence](images/ai_landscape.png)
+
+image source: [RapidOps](https://www.rapidops.com/blog/generative-ai-tools/)
+
+# Math
+
+## Math Topics
+
+:::: {.columns}
+
+::: {.column width="45%"}
+**Gradients** and their relation to local minima/maxima
+
+The **chain rule** for differentiation
+
+Matrices as **linear transformations** for vectors
+
+Notions of **basis**/rank/span/independence/etc.
+:::
+
+::: {.column width="10%"}
+
+:::
+
+::: {.column width="45%"}
+[calculus](https://www.youtube.com/watch?v=WUvTyaaNkzM&list=PLZHQObOWTQDMsr9K-rj53DwVRMYO3t5Yr)
+
+[linear algebra](https://www.youtube.com/watch?v=fNk_zzaMoSs&list=PLZHQObOWTQDPD3MizzM2xVFitgF8hE_ab)
+:::
+
+::::
+
+
+# Programming
+
+:::: {.columns}
+
+::: {.column width="45%"}
+Python
+
+C++
+
+CUDA
 :::
 
-::: notes
-- You can add notes on each slide with blocks like this!
-- Load a deck in the browser and type "s" to see these notes.
+::: {.column width="10%"}
+
 :::
 
-# SLIDE SECTION
+::: {.column width="45%"}
+![Python](images/python-logo-only.png)
+
+image source: [Python Software Foundation](https://www.python.org/community/logos/)
+:::
 
-## SLIDE
+::::
 
-- DENOTE MAJOR SECTIONS WITH `# TITLE` (eg `# Installation`)
-- ADD INDIVIDUAL SLIDES WITH `##` (eg `## rustup on Linux/macOS`)
-- KEEP THEM RELATIVELY SLIDE-LIKE; THESE ARE NOTES, NOT THE BOOK ITSELF.
 
-## SLIDE
 
-# SLIDE SECTION
 
-## SLIDE
 
-## SLIDE
diff --git a/slides/02_statistical-prediction-and-supervised-learning.qmd b/slides/02_statistical-prediction-and-supervised-learning.qmd
@@ -5,27 +5,82 @@ title: "2. Statistical Prediction and Supervised Learning"
 
 # Learning objectives
 
-::: nonincremental
-- THESE ARE NICE TO HAVE BUT NOT ABSOLUTELY NECESSARY
+:::: {.columns}
+
+::: {.column width="30%"}
+Review statistics terms
+
+Discuss machine learning ideas
+:::
+
+::: {.column width="10%"}
+
 :::
 
-::: notes
-- You can add notes on each slide with blocks like this!
-- Load a deck in the browser and type "s" to see these notes.
+::: {.column width="60%"}
+![Machine Learning](images/machine_learning_web.png)
+
+image source: [Sankalp Salve](https://medium.com/@xsankalp13/advanced-topics-in-machine-learning-e55d8b0dc7f9)
+:::
+
+::::
+
+# Statistics and Probability
+
+[![Random Variables](images/random_variables_3Blue1Brown.png)](https://www.youtube.com/watch?v=zeJD6dqJ5lo)
+
+# Machine Learning
+
+## Regression vs Classification
+
+![Regression or Classification](images/regression-vs-classification_simple-comparison-image_v3.png)
+
+image source: [Sharp Sight Labs](https://www.sharpsightlabs.com/blog/regression-vs-classification/)
+
+## Supervised Learning vs Unsupervised Learning
+
+![Supervised or Unsupervised](images/Supervised-learning-and-unsupervised-learning-Supervised-learning-uses-annotation.png)
+
+image source: [Ma Yan, et al](https://www.researchgate.net/figure/Supervised-learning-and-unsupervised-learning-Supervised-learning-uses-annotation_fig1_329533120)
+
+## Regularization
+
+::: incremental
+
+* Loss function $$L(\vec{\beta}) = \text{argmin}_{\vec{\beta}} \sum_{i=1}^{N} \left(y_{i} - \beta_{0} - \sum_{j=1}^{k} \beta_{j}x_{ij}\right)^{2}$$
+
+* L1 Regularization $$L(\vec{\beta}, \lambda) = \text{argmin}_{\vec{\beta}} \left[\sum_{i=1}^{N} \left(y_{i} - \beta_{0} - \sum_{j=1}^{k} \beta_{j}x_{ij}\right)^{2} + \lambda\sum_{j=i}^{k}|\beta_{j}|\right]$$
+
+* L2 Regularization $$L(\vec{\beta}, \lambda) = \text{argmin}_{\vec{\beta}} \left[\sum_{i=1}^{N} \left(y_{i} - \beta_{0} - \sum_{j=1}^{k} \beta_{j}x_{ij}\right)^{2} + \lambda\sum_{j=i}^{k}\beta_{j}^{2}\right]$$
+
 :::
 
-# SLIDE SECTION
+## Empirical Risk Minimization
+
+We can approximate the *expected risk* over a loss function, data set, and hypothesis model $h$ 
+
+$$\text{E}\left[L((\vec{x}, \vec{y}), h)\right]$$
+by taking the average over the training data
+
+$$\frac{1}{n}\displaystyle\sum_{i=1}^{n} L((x_{i}, y_{i}), h)$$
+* formulas outlined by [Professor Alexander Jung](https://www.youtube.com/watch?v=8fsJCyBOizQ)
+
+## Bias-Variance Tradeoff
+
+Within a *hypothesis class* of similar modeling functions, we are concerned with the **bias-variance tradeoff** in model selection.
+
+![bias-variance tradeoff](images/biasvariance.png)
+
+image source: [Scott Fortmann-Roe ](http://scott.fortmann-roe.com/docs/BiasVariance.html)
+
 
-## SLIDE
+## Linear Foundations
 
-- DENOTE MAJOR SECTIONS WITH `# TITLE` (eg `# Installation`)
-- ADD INDIVIDUAL SLIDES WITH `##` (eg `## rustup on Linux/macOS`)
-- KEEP THEM RELATIVELY SLIDE-LIKE; THESE ARE NOTES, NOT THE BOOK ITSELF.
+"For more complex and high-dimensional problems with potential nonlinear dependencies between features, it’s often useful to ask:
 
-## SLIDE
+* What’s a linear model for the problem?
+* Why does the linear model fail?
+* What’s the best way to add nonlinearity, given the semantic structure of the problem?
 
-# SLIDE SECTION
 
-## SLIDE
 
-## SLIDE
diff --git a/slides/03_time-series-analysis.qmd b/slides/03_time-series-analysis.qmd
@@ -6,26 +6,67 @@ title: "3. Time-Series Analysis"
 # Learning objectives
 
 ::: nonincremental
-- THESE ARE NICE TO HAVE BUT NOT ABSOLUTELY NECESSARY
+- Mention time series
 :::
 
-::: notes
-- You can add notes on each slide with blocks like this!
-- Load a deck in the browser and type "s" to see these notes.
-:::
+## Package Libraries
+
+```{r}
+#| message: false
+#| warning: false
+
+library("DiagrammeR")
+```
+
+
+## Stepping Forward
+
+```{r}
+#| echo: false
+#| eval: true
+DiagrammeR::mermaid("
+graph LR
+
+TS[time<br />series]
+ARIMA[autoregressive<br />models]
+TIS[time<br />invariant<br />systems]
+CT[control<br />theory]
+NSSM[neural<br />state space<br />models]
+
+TS --> ARIMA
+
+ARIMA --> diffusion
+ARIMA --> LLMs
 
-# SLIDE SECTION
+diffusion --> SDEs
+diffusion --> NSSM
+TIS --> NSSM
+CT --> NSSM
+")
+```
 
-## SLIDE
+## Mermaid code
 
-- DENOTE MAJOR SECTIONS WITH `# TITLE` (eg `# Installation`)
-- ADD INDIVIDUAL SLIDES WITH `##` (eg `## rustup on Linux/macOS`)
-- KEEP THEM RELATIVELY SLIDE-LIKE; THESE ARE NOTES, NOT THE BOOK ITSELF.
+```{r}
+#| echo: true
+#| eval: false
+DiagrammeR::mermaid("
+graph LR
 
-## SLIDE
+TS[time<br />series]
+ARIMA[autoregressive<br />models]
+TIS[time<br />invariant<br />systems]
+CT[control<br />theory]
+NSSM[neural<br />state space<br />models]
 
-# SLIDE SECTION
+TS --> ARIMA
 
-## SLIDE
+ARIMA --> diffusion
+ARIMA --> LLMs
 
-## SLIDE
+diffusion --> SDEs
+diffusion --> NSSM
+TIS --> NSSM
+CT --> NSSM
+")
+```
diff --git a/slides/_metadata.yml b/slides/_metadata.yml
@@ -4,7 +4,7 @@ format:
     footer: <a href="/" target="genai_club-index">DSLC.io/genai</a> | <a href="https://DSLC.io" target="_blank">DSLC.io</a>
     link-external-newwindow: true
     transition: slide
-    incremental: true
+    incremental: false
 execute:
   eval: false
   echo: true

diff --git a/...ised-learning-and-unsupervised-learning-Supervised-learning-uses-annotation.png b/...ised-learning-and-unsupervised-learning-Supervised-learning-uses-annotation.png
diff --git a/slides/images/ai_landscape.png b/slides/images/ai_landscape.png
diff --git a/slides/images/biasvariance.png b/slides/images/biasvariance.png
diff --git a/slides/images/gen_ai_splash.png b/slides/images/gen_ai_splash.png
diff --git a/slides/images/machine_learning_web.png b/slides/images/machine_learning_web.png
diff --git a/slides/images/python-logo-only.png b/slides/images/python-logo-only.png
diff --git a/slides/images/random_variables_3Blue1Brown.png b/slides/images/random_variables_3Blue1Brown.png
diff --git a/slides/images/regression-vs-classification_simple-comparison-image_v3.png b/slides/images/regression-vs-classification_simple-comparison-image_v3.png