From 6d521a140e4dfb1604acab3a358140cfba16f4f1 Mon Sep 17 00:00:00 2001
From: Matthew Feickert <matthew.feickert@cern.ch>
Date: Wed, 24 Jan 2024 12:32:46 -0600
Subject: [PATCH] Move active learning to backup and put down HS3

---
 talk.md | 78 +++++++++++++++++++++++++++++++--------------------------
 1 file changed, 42 insertions(+), 36 deletions(-)

diff --git a/talk.md b/talk.md
index ac178e2..8d146a3 100644
--- a/talk.md
+++ b/talk.md
@@ -900,12 +900,6 @@ Requires all operations to be differentiable
 - Analysis end-to-end optimized directly on physics sensitivity
 ]
 
----
-class: focus-slide, center
-# Scalable solutions
-
-.huge.bold.center[Differentiable analyses at LHC scale]
-
 ---
 # Scaling is reasonable
 
@@ -957,15 +951,16 @@ From the 2023 MIAPbP Workshop on on Differentiable and Probabilistic Programming
    </a>
 </p>
 
-.center[Revisiting IRIS-HEP Analysis Systems in the context of distributed scaling and analysis reuse]
+.center[Revisiting .bold[IRIS-HEP Analysis Systems] in the context of distributed scaling and analysis reuse]
 
 ---
 # Analysis Reuse
 
 .large[
-* Data and analyses done at the LHC are unique physics opportunities
-* [RECAST](https://github.com/recast-hep/recast-atlas) has been implemented in ATLAS as an enabling technology
-* Resulting in ATLAS PUB notes extending the physics reach of original publications
+* Data and analyses done at the LHC are .bold[unique scientific opportunities]
+   - They can, and only ever will be, done at one place (LHC) at one time (now!).
+* [RECAST](https://github.com/recast-hep/recast-atlas) .bold[reinterpretation and analysis preservation] framework (Heinrich, Feickert) has been implemented in ATLAS as an enabling technology
+* Results leading to ATLAS public notes extending the physics reach of original publications
 ]
 
 .kol-1-3[
@@ -987,34 +982,27 @@ From the 2023 MIAPbP Workshop on on Differentiable and Probabilistic Programming
 .kol-1-3[
 <p style="text-align:center;">
    <a href="https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2021-020/">
-      <img src="figures/ATL-PHYS-PUB-2021-020.png"; width=90%>
+      <img src="figures/ATL-PHYS-PUB-2021-020.png"; width=88%>
    </a>
 </p>
 .caption[[ATL-PHYS-PUB-2021-020](https://inspirehep.net/literature/1870397)]
 ]
 
 ---
-# ML + reinterpretation: Active learning
+# Publishing statistical models
+
+.bold[TODO: Add slide about impact of pyhf]
 
-.kol-1-2[
-.huge[
-Leveraging [REANA](https://www.reana.io/) reproducible research data analysis platform possible to run distributed ML and analysis workflows at scale
-]
-<p style="text-align:center;">
-   <a href="https://conference.ippp.dur.ac.uk/event/1178/contributions/6449/">
-      <img src="figures/exclusion-learning.png"; width=90%>
-   </a>
-</p>
-.caption[[ Christian Weber, Reinterpretation Forum 2023](https://conference.ippp.dur.ac.uk/event/1178/contributions/6449/)]
-]
-.kol-1-2[
-<p style="text-align:center;">
-   <a href="https://conference.ippp.dur.ac.uk/event/1178/contributions/6449/">
-      <img src="figures/active-learning-workflow.png"; width=100%>
-   </a>
-</p>
-.caption[[ATL-PHYS-PUB-2023-010](https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2023-010/)]
-]
+
+---
+# Publishing statistical model standards
+
+.bold[TODO: Add slide about HS3]
+
+---
+# Dedicated investment for the next decade
+
+.bold[TODO: Add slide about P5 and funding for the next decade]
 
 ---
 # Applications beyond HEP
@@ -1028,11 +1016,6 @@ Leveraging [REANA](https://www.reana.io/) reproducible research data analysis pl
 * Planning for analysis reuse brings flexibility to leverage tooling
 ]
 
----
-# Dedicated support for the next decade
-
-.bold[TODO: Add slide about P5 and funding for the next decade]
-
 ---
 # Summary
 
@@ -1337,6 +1320,29 @@ Please checkout the many resources we have starting with the [website](https://s
 .bold.center[Groups, libraries, and applications growing rapidly]
 ]
 
+---
+# ML + reinterpretation: Active learning
+
+.kol-1-2[
+.huge[
+Leveraging [REANA](https://www.reana.io/) reproducible research data analysis platform possible to run distributed ML and analysis workflows at scale
+]
+<p style="text-align:center;">
+   <a href="https://conference.ippp.dur.ac.uk/event/1178/contributions/6449/">
+      <img src="figures/exclusion-learning.png"; width=90%>
+   </a>
+</p>
+.caption[[ Christian Weber, Reinterpretation Forum 2023](https://conference.ippp.dur.ac.uk/event/1178/contributions/6449/)]
+]
+.kol-1-2[
+<p style="text-align:center;">
+   <a href="https://conference.ippp.dur.ac.uk/event/1178/contributions/6449/">
+      <img src="figures/active-learning-workflow.png"; width=100%>
+   </a>
+</p>
+.caption[[ATL-PHYS-PUB-2023-010](https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2023-010/)]
+]
+
 ---
 # Active Learning