AIES submission done

Author: pat-alt

dev/submissions/aies/extended_abstract/extended_abstract.fdb_latexmk

@@ -1,6 +1,6 @@
 # Fdb version 3
-["pdflatex"] 1651243546 "/Users/FA31DU/Documents/code/algorithmic_recourse_dynamics/dev/submissions/aies/extended_abstract/extended_abstract.tex" "extended_abstract.pdf" "extended_abstract" 1651243549
-  "/Users/FA31DU/Documents/code/algorithmic_recourse_dynamics/dev/submissions/aies/extended_abstract/extended_abstract.tex" 1651243544 17053 6fe9519a7d0eef8e644ca6e87602ba81 ""
+["pdflatex"] 1651317361 "/Users/FA31DU/Documents/code/algorithmic_recourse_dynamics/dev/submissions/aies/extended_abstract/extended_abstract.tex" "extended_abstract.pdf" "extended_abstract" 1651317363
+  "/Users/FA31DU/Documents/code/algorithmic_recourse_dynamics/dev/submissions/aies/extended_abstract/extended_abstract.tex" 1651317347 17039 33175e48a0264725ef3fdaa9539a2d61 ""
   "/usr/local/texlive/2022/texmf-dist/fonts/enc/dvips/lm/lm-ec.enc" 1254269338 2375 baa924870cfb487815765f9094cf3728 ""
   "/usr/local/texlive/2022/texmf-dist/fonts/enc/dvips/lm/lm-mathit.enc" 1202520719 2405 5dcf2c1b967ee25cc46c58cd52244aed ""
   "/usr/local/texlive/2022/texmf-dist/fonts/enc/dvips/lm/lm-mathsy.enc" 1202520719 2840 216e6e45ad352e2456e1149f28885bee ""
@@ -287,8 +287,8 @@
   "/usr/local/texlive/2022/texmf-var/fonts/map/pdftex/updmap/pdftex.map" 1647878959 4410336 7d30a02e9fa9a16d7d1f8d037ba69641 ""
   "/usr/local/texlive/2022/texmf-var/web2c/pdftex/pdflatex.fmt" 1649828967 2826441 25557c603d1b120c5ae526d9d48800f2 ""
   "/usr/local/texlive/2022/texmf.cnf" 1647878952 577 209b46be99c9075fd74d4c0369380e8c ""
-  "extended_abstract.aux" 1651243548 5146 a3c21ced8c7b7c8c5af8d187c3b1d3c2 "pdflatex"
-  "extended_abstract.tex" 1651243544 17053 6fe9519a7d0eef8e644ca6e87602ba81 ""
+  "extended_abstract.aux" 1651317363 5146 a3c21ced8c7b7c8c5af8d187c3b1d3c2 "pdflatex"
+  "extended_abstract.tex" 1651317347 17039 33175e48a0264725ef3fdaa9539a2d61 ""
   "www/poc.png" 1651241025 59205 e5d353bb5bcd947e141460427face064 ""
   (generated)
   "extended_abstract.aux"

dev/submissions/aies/extended_abstract/extended_abstract.pdf

@@ -12,7 +12,7 @@ format:
 
 ## Introduction
 
-Recent advances in artificial intelligence (AI) have propelled its adoption in domains outside of computer science including health care, bioinformatics and genetics. In finance, economics and other social sciences, applications of AI are still relatively limited. Decision-making in these fields has traditionally been guided by interpretable models that facilitate explanations. Explainability is crucial in this context, since decision-makers are typically held accountable by the public: central banks, for example, are heavily scrutinized for the policies they impose. It is therefore not surprising that practitioners and academics in these fields are reluctant to adopt AI technologies they cannot trust. Deep neural networks, for example, are generally considered as black boxes and therefore not trustworthy in a context that demands explanations. This PhD project is focused on exploring and developing methodologies that improve the trustworthiness of AI and thereby enable its application in Finance and Economics. 
+Recent advances in Artificial Intelligence (AI) have propelled its adoption in domains outside of Computer Science including Healthcare, Bioinformatics and Genetics. In Finance, Economics and other social sciences, applications of AI are still relatively limited. Decision-making in these fields has traditionally been guided by interpretable models that facilitate explanations. Explainability is crucial in this context, since decision-makers are typically held accountable by the public: central banks, for example, are heavily scrutinized for the policies they impose. It is therefore not surprising that practitioners and academics in these fields are reluctant to adopt AI technologies they cannot trust. Deep neural networks, for example, are generally considered as black boxes and therefore not trustworthy in a context that demands explanations. This PhD project is focused on exploring and developing methodologies that improve the trustworthiness of AI and thereby enable its application in Finance and Economics. 
 
 The remainder of this extended abstract is structured as follows: @sec-main presents one of the research questions I have investigated during the first months of my PhD: how do counterfactual explanations handle dynamics? @sec-related places this work in the broader context of my research.
 
@@ -26,7 +26,7 @@ This project investigates **endogenous** domain and model shifts, that is shifts
 
 These dynamics may be problematic. As the decision boundary moves in the direction of the non-target class, counterfactual paths become shorter: in the loan example, individuals that previously would have been denied credit based on their input features are suddenly considered as creditworthy. Average default risk across all borrowers can therefore be expected to increase. Conversely, lenders that anticipate such dynamics may choose to deny credit to individuals that have implemented AR, thereby compromising the validity of AR. 
 
-To the best of my knowledge this is the first work investigating endogenous dynamic in AR. Through future experiments I want to investigate how this phenomenon plays out across different benchmark datasets including German credit, Boston Housing and COMPAS.^[These benchmark datasets have their issues and controversies, which is one of the challenges I would like to discuss at AIES.] Furthermore, I want to assess to what extent the magnitude and direction of domain and model shifts depends on the choice of the counterfactual generator. To this end, I am currently supervising a group of undergraduate students, who are tackling some of these tasks in their final-year research project. 
+To the best of my knowledge this is the first work investigating endogenous dynamics in AR. Through future experiments I want to investigate how this phenomenon plays out across different benchmark datasets including German credit, Boston Housing and COMPAS.^[These benchmark datasets have their issues and controversies, which is one of the challenges I would like to discuss at AIES.] Furthermore, I want to assess to what extent the magnitude and direction of domain and model shifts depends on the choice of the counterfactual generator. To this end, I am currently supervising a group of undergraduate students, who are tackling some of these tasks in their final-year research project. 
 
 ![Dynamics in Algorithmic Recourse: we have a simple Bayesian model trained for binary classification (a); the implementation of AR for a random subset of individuals leads to a domain shift (b); as the classifier is retrained we observe a model shift (c); as this process is repeated, the decision boundary moves away from the target class (d).](www/poc.png){#fig-dynamics fig.pos="h" width=45%}
 

dev/submissions/aies/extended_abstract/extended_abstract.qmd

@@ -186,19 +186,19 @@
 \makeatother
 \subtitle{Trustworthy Artificial Intelligence for Finance and Economics}
 \author{}
-\date{April 29, 2022}
+\date{}
 
 \begin{document}
 \maketitle
 
-\ifdefined\Shaded\renewenvironment{Shaded}{\begin{tcolorbox}[boxrule=0pt, sharp corners, frame hidden, borderline west={3pt}{0pt}{shadecolor}, interior hidden, enhanced, breakable]}{\end{tcolorbox}}\fi
+\ifdefined\Shaded\renewenvironment{Shaded}{\begin{tcolorbox}[sharp corners, breakable, borderline west={3pt}{0pt}{shadecolor}, interior hidden, frame hidden, enhanced, boxrule=0pt]}{\end{tcolorbox}}\fi
 
 \hypertarget{introduction}{%
 \section{Introduction}\label{introduction}}
 
-Recent advances in artificial intelligence (AI) have propelled its
-adoption in domains outside of computer science including health care,
-bioinformatics and genetics. In finance, economics and other social
+Recent advances in Artificial Intelligence (AI) have propelled its
+adoption in domains outside of Computer Science including Healthcare,
+Bioinformatics and Genetics. In Finance, Economics and other social
 sciences, applications of AI are still relatively limited.
 Decision-making in these fields has traditionally been guided by
 interpretable models that facilitate explanations. Explainability is
@@ -270,7 +270,7 @@ \section{Dynamics in Algorithmic Recourse}\label{sec-main}}
 thereby compromising the validity of AR.
 
 To the best of my knowledge this is the first work investigating
-endogenous dynamic in AR. Through future experiments I want to
+endogenous dynamics in AR. Through future experiments I want to
 investigate how this phenomenon plays out across different benchmark
 datasets including German credit, Boston Housing and COMPAS.\footnote{These
   benchmark datasets have their issues and controversies, which is one

dev/submissions/aies/extended_abstract/extended_abstract.synctex.gz

@@ -3,7 +3,7 @@ format:
     pdf: default
 ---
 
-## Personal details
+## Personal Details
 
 |  | Author | Research Advisor |
 |---------|:--------|:--------|

dev/submissions/aies/extended_abstract/extended_abstract.tex

@@ -1,12 +1,13 @@
 ---
+title: Personal Statement
 format:
     pdf: default
 ---
 
 ## Personal Statement
 
-As a first-year PhD student in computer science with a background in finance and economics, I was very happy to see that AIES offers a student track. I have worked on several projects during the first months of my PhD and while the one I hope to present at AIES is the most mature research project at this point, it is still very much a work in progress. I think that the question I pose about the dynamics of algorithmic recourse is an interesting, albeit still fairly broad and open one. Through my participation in AIES, I would hope to get feedback that helps me refine the research question. As experiments related to the question are now being carried out by my students, the work can be expected to progress further until the actual AIES event and hence I could share more experimental results that may be of interest to the broader community. AIES would also be an opportunity to discuss ideas for potential remedies to the issues this work highlights. 
+As a first-year PhD student in Computer Science with a background in Finance and Economics, I was very happy to see that AIES offers a student track. I have worked on several projects during the first months of my PhD and while the one I hope to present at AIES is the most mature research project at this point, it is still very much a work-in-progress. I think that the question I pose about the dynamics in Algorithmic Recourse is an interesting, albeit still fairly broad and open one. Through my participation in AIES, I would hope to get feedback that helps me refine the research question. As experiments related to the question are now being carried out by a group of students under my supervision, the work can be expected to progress further until the actual AIES event. I therefore expect that I can share more experimental results that may be of interest to the broader community. AIES would also be an opportunity to discuss ideas for potential remedies to the issues this work highlights. 
 
-I also think that my broader PhD research agenda on the intersection of trustworthy AI and the social sciences should be of interest to the AIES community. Thanks to my academic background in economics, finance and data science as well my previous professional experience in monetary policy, I believe that I can contribute insights to a wide range of discussions revolving around the diverse set of topics relevant to AIES. It would be immensely helpful to learn from more experienced colleagues as well as fellow early-stage researchers, who are working on related research questions. At the main conference and the student event I expect to be exposed to new research ideas that I have never thought off, but may well be highly relevant to my own research. I would even hope to potentially find opportunities for future collaborations.
+I also think that my broader PhD research agenda on the intersection of trustworthy AI and the social sciences should be of interest to the AIES community. Thanks to my academic background in Economics, Finance and Data Science as well my previous professional experience in Monetary Policy, I believe that I can contribute insights to a wide range of discussions revolving around the diverse set of topics relevant to AIES. It would be immensely helpful to learn from more experienced colleagues as well as fellow early-stage researchers, who are working on related research questions. At the main conference and the student event I expect to be exposed to new research ideas that I have not thought of myself, but that may well be highly relevant to my own research. I would even hope to potentially find opportunities for future collaborations.
 
-This would be the first student program I have attended, although it is worth mentioning that in the past I have presented research at the European Central Bank (2019), IFABS (2021), NeurIPS (2021) and the Irving Fisher Committee on Central Bank Statistics (2022). Since the main PhD research project that I would like to dicuss is still in its early stages, I have not submitted anything to the main track of AIES. I would hope that following my participation in the student track I can use the feedback to carry the work forward and submit it to the main track next year. 
+This would be the first student program I have attended, although it is worth mentioning that in the past I have presented research at the European Central Bank (2019), Bank of England, IFABS (2021), NeurIPS (2021) and the Irving Fisher Committee on Central Bank Statistics (2022). More informations on this can be found in my resume. Since the main PhD research project that I would like to dicuss is still in its early stages, I have not submitted anything to the main track of AIES. I would hope that following my participation in the student track I can use the feedback to carry the work forward and submit it to the main track next year.