vault backup: 2023-06-12 19:07:33

Affected files:
.obsidian/backlink.json
.obsidian/graph.json
.obsidian/workspace-mobile.json
.obsidian/workspace.json
Events/🪣🪣🪣.md
Health/ADHD.md
STEM/AI/Classification/Gradient Boosting Machine.md
STEM/AI/Neural Networks/CV/Visual Search/Visual Search.md
STEM/AI/Neural Networks/Learning/Tasks.md
STEM/AI/Pattern Matching/Dynamic Time Warping.md
STEM/AI/Problem Solving.md
STEM/CS/Regex.md
STEM/img/dtw-graph-unit.png
STEM/img/dtw-graph.png
STEM/img/dtw-gross-partitioning.png
STEM/img/dtw-heatmap-distortion.png
STEM/img/dtw-heatmap.png
STEM/img/dtw-possible-movements.png
STEM/img/dtw-score-pruning.png
STEM/img/nn-tasks-function-approx-inverse.png
STEM/img/nn-tasks-function-approx.png
STEM/img/nn-tasks-pattern.png
STEM/img/problem-solving-arch.png
STEM/img/problem-solving-goal-based.png
STEM/img/problem-solving-reflex.png
STEM/img/visual-search-arch.png
STEM/img/visual-search-crude.png

AI/Classification/Gradient Boosting Machine.md

@@ -5,3 +5,4 @@
 	- Similar properties
 - One of the best algorithm for dealing with non perceptual data
 - XGBoost
+	- [Confusing XGBoost Hyperparameters](https://towardsdatascience.com/10-confusing-xgboost-hyperparameters-and-how-to-tune-them-like-a-pro-in-2023-e305057f546?source=rss----7f60cf5620c9---4)

AI/Neural Networks/CV/Visual Search/Visual Search.md

@@ -0,0 +1,10 @@
+- Shallow would be BOVW
+- Use metric space over feature space
+	- Get ranked list
+
+![](../../../../img/visual-search-arch.png)
+# Crude Method
+- Doesn't enforce metric relationship
+- Sample prior to final softmax
+![](../../../../img/visual-search-crude.png)
+

AI/Neural Networks/Learning/Tasks.md

@@ -0,0 +1,44 @@
+# Pattern Association
+- Associative memory
+	- Learns by association
+- Autoassociation
+	- Store a set of patterns by repeatedly presenting them in the network
+		- Then presented partial or distorted stored pattern
+		- Recall intended
+	- Input and output data spaces are same dimensionality
+- Heteroassociation
+	- Arbitrary set of input patterns paired with another arbitrary set of output patterns
+		- Supervised instead of unsupervised
+	- No required relationship between input/output dimensionality
+- Stages
+	- Storage
+	- Recall
+
+# Pattern Recognition
+- Received pattern/signal is assigned to one of a prescribed number of classes
+![](../../../img/nn-tasks-pattern.png)
+
+# Function Approximation
+- System Identification
+![](../../../img/nn-tasks-function-approx.png)
+- Inverse System
+![](../../../img/nn-tasks-function-approx-inverse.png)
+
+# Control
+- Learn to control a process or critical part of a system
+
+# Filtering
+- Filtering
+	- Extraction of information about a quantity of interest at discrete time $n$ by using data from time up to $n$
+- Smoothing
+	- Use information past time $n$
+		- Expect smoother result
+	- Delay in processing
+- Prediction
+	- Predict later data using current and previous
+
+# Beamforming
+- Spatial filtering
+- Distinguish spatial properties of a target signal and background noise
+	- Similar to bats
+	- Used in radar and sonar

AI/Pattern Matching/Dynamic Time Warping.md

@@ -0,0 +1,73 @@
+***Deterministic
+Pattern Recogniser***
+Allows timescale variations in sequences for same class
+
+![](../../img/dtw-graph.png)
+
+$$D(T,N)=\min_{t,i}\sum_{\substack{t\in1..T \\ i\in1..N}}d(t,i)$$
+- $d(t,i)$ is distance between features from $t$-th frame of test to $i$-th frame of template
+
+$$D(t,i)=\min[D(t,i-1),D(t-1, i-1),D(t-1,i)]+d(t,i)$$
+- Allowing transition from current and previous frame only
+- Recursive
+
+![](../../img/dtw-graph-unit.png)
+
+# Problems
+- How much flexibility to allow?
+- How to penalise warping?
+- How to determine a fair distance metric?
+- How many templates to register?
+	- How to select best ones?
+
+# Basic Algorithm
+1. Initialise the cumulative distances for $t=1$
+$$D(1,i)=\begin{cases}d(1,i) & \text{for }i=1, \\ D(1, i-1)+d(1,i) & \text{for }i=2,...,N\end{cases}$$
+2. Recur for $t=2,...,T$
+$$D(t,i)=\begin{cases}D(t-1,i) + d(t,i) & \text{for }i=1, \\ \min[D(t, i-1), D(t-1, i-1),D(t-1,i)] + d(t,i) & \text{for }i=2,...,N\end{cases}$$
+3. Finalise, the cumulative distance up to the final point gives the total cost of the match: $D(T,N)$
+
+![](../../img/dtw-heatmap.png)
+- Euclidean distances
+
+# Distortion Penalty
+1. Initialise the cumulative distances for $t=1$
+$$D(1,i)=\begin{cases}d(1,i) & \text{for }i=1, \\ d(1,i)+D(1, i-1)+d_V & \text{for }i=2,...,N\end{cases}$$
+2. Recur for $t=2,...,T$
+$$D(t,i)=\begin{cases}d(t,i)+D(t-1,i1)+d_H & \text{for }i=1, \\ \min[d(t,i)+D(t,i-1)+d_V,2d(t,i)+D(t-1,i-1),d(t,i)+D(t-1,i)+d_H] & \text{for }i=2,...,N\end{cases}$$
+- Where $d_V$ and $d_H$ are costs associated with vertical and horizontal transitions respectively
+3. Finalise, the cumulative distance up to the final point gives the total cost of the match: $D(T,N)$
+
+- Allows weighting for dynamic penalties when moving horizontally or vertically
+	- As opposed to diagonally
+
+![](../../img/dtw-heatmap-distortion.png)
+
+# Store Best Path
+1. Initialise distances and traceback indicator for $t=1$
+$$D(1,i)=\begin{cases}d(1,i) & \text{for } i=1,\\ d(1,i)+D(1,i-1) & \text{for }i = 2,...,N\end{cases}$$
+$$\phi(1,i)=\begin{cases}[0,0] & \text{for } i=1,\\ [1,i-1] & \text{for }i = 2,...,N\end{cases}$$
+2. Recur for cumulative distances at $t=2,...,T$
+$$D(1,i)=\begin{cases}d(t,i)+D(t-1,i) & \text{for } i=1,\\ d(t,i)+\min[D(t,i-1),D(t-1,i-1),D(t-1,i)] & \text{for }i = 2,...,N\end{cases}$$
+$$\phi(1,i)=\begin{cases}[t-1,i] & \text{for } i=1,\\ \arg\min[D(t,i-1),D(t-1,i-1),D(t-1,i)] & \text{for }i = 2,...,N\end{cases}$$
+3. Final point gives the total alignment cost D(T,N) and the end coordinates of the best path $z_K=[T,N]$, where $K$ is the number of nodes on the optimal path
+4. Trace the path back for $k=K-1,...,1,z_k=\phi(z_{k+1}), \text{ and }Z=\{z_1,...,z_K\}$
+- Stores best path
+
+![](../../img/dtw-possible-movements.png)
+- Vary allowable movements through grid
+- Second row for blocking multiple of the same movements in succession
+
+# Search Pruning
+- Speed up algorithm for real-time
+- Kill bad options
+
+## Gross Partitioning
+![](../../img/dtw-gross-partitioning.png)
+- Too far from diagonal
+- Probably wrong or bad
+
+## Score Pruning
+![](../../img/dtw-score-pruning.png)
+- Examine existing branches
+- See which scores are really bad

AI/Problem Solving.md

@@ -0,0 +1,57 @@
+# Problem Types
+- Toy/game problems
+	- Illustrative
+- Real world problems
+	- Exact solutions
+
+# Intelligent Agents
+*Any thing which can be viewed as perceiving its environment through sensors and acting upon that environment through its affectors*
+
+## Reflex
+![](img/problem-solving-reflex.png)
+## Goal Based
+![](img/problem-solving-goal-based.png)
+## Utility-based
+
+# Environments
+- (in)accessible
+- (non)-deterministic
+- (non)-episodic
+- Static/dynamic
+- Discrete/continuous
+
+# Solving
+1. Precise problem definition
+	- Start conditions and goal state
+2. Problem analysis
+	- Appropriate representation
+	- Problem characteristics
+		- Knowledge intensive
+		- Decomposability
+3. Choice of best technique
+
+# Moves
+- Ignorable
+	- Simple control structure
+- Recoverable
+	- Backtrack using control stack
+- Irrecoverable
+	- Much effort needed for each decision/move
+
+![](img/problem-solving-arch.png)
+- Control
+	- Cause movement
+	- Be systematic
+	- Be guided by heuristics
+
+# Predictability
+## 8 Puzzle
+- Can plan ahead with certainty — only need a   
+control structure with backtracking   
+	- recoverable, certain outcome  
+## Chess
+- Future moves are not predictable with certainty.   
+- May need to consider several possibilities, and decide best option based on Incomplete information   
+- irrecoverable — once a piece is taken cannot in general recover from this state.
+- deciding on an optimal move can be difficult
+- try to predict opponent's move

CS/Regex.md

@@ -0,0 +1,14 @@
+# Markdown
+
+## All links
+`\[.*?\]\(.*?\)`
+
+## Images
+`!\[.*?\]\(.*?\.png\)`
+
+## Inter-Ref
+`\[.*?\]\(.*?\.md(#.*?)*\)`
+
+## Tags
+`#{1}[^\s#.]+`
+`(^|[[:blank:]])#{1}[^\s#.]+`