notxia/unibo-ai-notes
1
0
Fork 0
mirror of https://github.com/NotXia/unibo-ai-notes.git synced 2025-12-15 02:52:22 +01:00
Code Issues Packages Projects Releases Wiki Activity

Add FAIKR decision support systems

Browse Source
This commit is contained in:
Tian Cheng (Riccardo) Xia
2023-09-30 11:12:43 +02:00
parent 8e0600375b
commit e31f9f8e69
2 changed files with 42 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
src/fundamentals-of-ai-and-kr/main.tex
View File

@ -8,8 +8,8 @@
\usepackage[inline]{enumitem}
\usepackage{marginnote}
\usepackage[bottom]{footmisc}
\geometry{ margin=3cm, lmargin=2cm, rmargin=4cm, marginparwidth=3cm }
\usepackage{scrlayer-scrpage}
\geometry{ margin=3cm, lmargin=1.5cm, rmargin=4.5cm, marginparwidth=3cm }
\hypersetup{ colorlinks, citecolor=black, filecolor=black, linkcolor=black, urlcolor=black, linktoc=all }
\NewDocumentEnvironment{descriptionlist}{}{%
@ -19,6 +19,7 @@
}
\setlength{\parindent}{0pt}
\renewcommand*{\marginfont}{\color{gray}\footnotesize}
\renewcommand*\chapterpagestyle{scrheadings} % Header in chapter pages
\theoremstyle{definition}
\newtheorem{theorem}{Theorem}[section]

43
src/fundamentals-of-ai-and-kr/sections/_intro.tex
View File

@ -1,3 +1,5 @@
\lohead{\color{gray} Not required for the exam}
\lehead{\color{gray} Not required for the exam}
\chapter{Introduction}
@ -71,7 +73,6 @@ Intelligence is defined as the ability to perceive or infer information and to r
\end{description}
\section{Machine learning}
\subsection{Training approach}
@ -136,14 +137,14 @@ The expressivity of a neural networks increases when more neurons are used:
Able to separate a convex region ($n_\text{edges} \leq n_\text{hidden neurons}$)
\begin{figure}[h]
\centering
\includegraphics[width=0.75\textwidth]{img/3layer.png}
\includegraphics[width=0.90\textwidth]{img/3layer.png}
\caption{Separation performed by a three-layer network}
\end{figure}
\item[Four-layer network]
Able to separate regions of arbitrary shape.
\begin{figure}[h]
\centering
\includegraphics[width=0.30\textwidth]{img/4layer.png}
\includegraphics[width=0.40\textwidth]{img/4layer.png}
\caption{Separation performed by a four-layer network}
\end{figure}
\end{descriptionlist}
@ -177,4 +178,38 @@ An \textbf{automated planner} is an agent that operates in a given domain descri
\section{Swarm intelligence}
\marginnote{Swarm intelligence}
Decentralized and self-organized systems that result in emergent behaviors.
Decentralized and self-organized systems that result in emergent behaviors.
\section{Decision support systems}
\begin{description}
\item[Knowledge based system] \marginnote{Knowledge based system}
Use knowledge (and data) to support human decisions.
Bottlenecked by knowledge acquisition.
\end{description}
Different levels of decision support exist:
\begin{descriptionlist}
\item[Descriptive analytics] \marginnote{Descriptive analytics}
Data are used to describe the system (e.g. dashboards, reports, \dots).
Human intervention is required.
\item[Diagnostic analytics] \marginnote{Diagnostic analytics}
Data are used to understand causes (e.g. fault diagnosis)
Decisions are made by humans.
\item[Predictive analytics] \marginnote{Predictive analytics}
Data are used to predict future evolutions of the system.
Uses machine learning models or simulators (digital twins)
\item[Prescriptive analytics] \marginnote{Prescriptive analytics}
Make decisions by finding the preferred scenario.
Uses optimization systems, combinatorial solvers or logical solvers.
\end{descriptionlist}
\newpage
\lohead{}
\lehead{}