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
This commit is contained in:
andy 2023-06-12 19:07:33 +01:00
parent 3606944190
commit b24f551589
21 changed files with 200 additions and 1 deletions

View File

@ -5,3 +5,4 @@
- Similar properties - Similar properties
- One of the best algorithm for dealing with non perceptual data - One of the best algorithm for dealing with non perceptual data
- XGBoost - 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)

View File

@ -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)

View File

@ -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

View File

@ -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

57
AI/Problem Solving.md Normal file
View File

@ -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

14
CS/Regex.md Normal file
View File

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

BIN
img/dtw-graph-unit.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.9 KiB

BIN
img/dtw-graph.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 55 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 11 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 512 KiB

BIN
img/dtw-heatmap.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 455 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 31 KiB

BIN
img/dtw-score-pruning.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 65 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 27 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 27 KiB

BIN
img/nn-tasks-pattern.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 34 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 36 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 32 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 20 KiB

BIN
img/visual-search-arch.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 34 KiB

BIN
img/visual-search-crude.png Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 69 KiB