PHYS 2801 Final Nomenclature

Operating System

Software that manages hardware and provides services for applications.

Terminal

Text-based interface to interact with the operating system.

Shell scripting

Writing sequences of commands in a file to automate tasks.

Virtual machine

Software emulation of a computer system

code repository

Central storage for code, often with version control

compiled code

Source code translated into machine-readable instructions

executable code

program file that can be run directly by the computer, not translated

LaTeX

Typesetting system for scientific and mathematical documents

Markdown

Lightweight markup language for formatting text

Python lists

Ordered, mutable collections of elements

Numpy arrays

Efficient, fixed-type, multidimensional arrays for numerical computing

Python methods (def)

User-defined functions that encapsulate reusable code

Numerical representation

Approximation of values using finite numbers

Analytical representation

Exact mathematical expression of a quantity

Numerical derivative

Approximation of a derivative using discrete data points

Numerical integral

Approximation of an integral using summation methods

Standard deviation

Measure of spread of data around the mean

Error propagation

How uncertainties in inputs affect outputs

Variance

Average squared deviation from the mean

Covariance

Measure of how 2 variables vary together

Interpolation

Estimating values between known data points

Grid search method

Exhaustive search over parameter space

Linear regression

Fitting a straight line to data

Method of least squares

Minimizing squared differences between data and model

Chi-squared

Statistic measuring deviation between observed and expected data

p-value

Probability of observing data as extreme as measured, assuming null hypothesis

Gradient descent

Iterative optimization method using slope information

First order ODE

Differential equation involving first derivatives

Second order ODE

Differential equation involving second derivatives

Euler's method

Simple numerical method for solving ODEs

Runge-Kutta

More accurate numerical method for solving ODEs

Why use code repositories?

Enable collaboration, version control, and backup

Git - clone

Copy a repository to your local machine

Git - add

Stage changes for commit

Git - push

Upload local commits to remote repository

Git - commit

Save changes with a message in version history

Git branches

Parallel version of code for development

probability vs statistics

Probability predicts outcomes, stats analyze observed data

Conditional probability

Probability of an event given another event

Independent events

Events whose outcomes do not affect each other

PDFs

Functions describing likelihood of continuous outcomes

Cumulative function

Probability of a variable that is less than or equal to a value

Quantile - percentile

Value below which a given percentage of data falls

Quantile - quartile

Division of data into 4 equal parts

Quantile - median

Middle value of ordered data

Expectation value (mean)

Average outcome weighted by probability

Correlation factors

Measure of linear relationship between variables

Random vs systematic errors

Random = unpredictable fluctuations; systematic = consistent bias

Binomial distribution

Probability distribution of successes in fixed trials

Poisson distribution

Probability of rare events in fixed intervals

Uniform distribution

Equal probability for all outcomes

Exponential distribution

Continuous probability distribution for time between events

Gaussian distribution

Bell-shaped distribution devined by mean and variance

Chi-squared PDF

Distribution of sum of squared normal variables

Principle of scientific method

Systematic process of hypothesis, experiment, and analysis

Quantitative vs qualitative

Numerical vs descriptive data

Hypothesis test

Statistical test of a proposed explanation

Post-hoc fallacy

Mistaken belief that correlation = causation

Directionality problem

Uncertainty about cause-effect direction

Confirmation bias

Tendency to favour evidence supporting beliefs

Cofounding factors

Inadequately measured 3rd variable distorts result of experiment

Systematic uncertainty sources

Biases from measurement or design that are consistent and repeatable

P-hacking

Manipulating data to achieve statistical significance

Blinding

Concealing information to reduce bias

Fixed methodology

Predefined procedures to ensure consistency

Artificial intelligence

Computer systems performing tasks requiring human-like intelligence

Machine learning

Algorithms that learn patterns from data

ML model

mathematical representation trained to map inputs to outputs

Artificial neuron

Basic unit computing weighted sum + activation

NN weights

Parameters scaling inputs

NN bias

Offset added to weighted sum

Layers of a NN

Stacked groups of artificial neurons

Deep network

Neural network with many layers

Dense network

Fully connected neural network (everything connected to everything)

Sigmoid activation function

Maps input to range (0,1)

Logistic regression in NN

Binary classification using sigmoid

Supervised learning

Training with labeled data

Unsupervised learning

Training without labels, finding structure

Loss function

Quantifies error between prediction and target

Epoch

One full pass through training data

PyTorch

ML framework for building neural networks

Keras

High-level API (application programming interface) for neural networks

LLMs

Neural networks trained on massive text datasets

ls

list files in directory

cd

change directory

pwd

print current working directory

date

display system date/time

hostname

show computer's network name

cp

copy files

rm

remove files

mv

move or rename files

wget

download files from the web

mkdir

create new directory

rmdir

remove directory

Use of pipes |

pass output of one command as input to another