Machinelearning

Abbiamo trattato i modelli classici in Convolutional Neural Network. Con i vecchi files di notion Il Kernel I punti interessanti delle immagini sono solamente i punti di cambio solo che attualmente siamo in stato discreto, quindi ci è difficile usare una derivata, si usano kernel del tipo: $\left[ 1, 0, -1 \right]$, che sarà positivo se cresce verso sinistra, negativo se scende. feature map Sono delle mappe che rappresentano alcune informazioni interessanti della nostra immagine. ...

Anomaly detection is a problem in machine learning that is of a big interest in industry. For example a bank needs to identify problems in transactions, doctors need it to see illness, or suspicious behaviors for law (no Orwell here). The main difference between this and classification is that here we have no classes. Setting of the problem Let’s say we have a set $X = \left\{ x_{1}, \dots, x_{n} \right\} \subseteq \mathcal{N} \subseteq \mathcal{X} = \mathbb{R}^{d}$ We say this set is the normal set, and $X$ are our samples but it’s quite complex, so we need an approximation to say whether if a set is normal or not. We need a function $\phi : \mathcal{X} \to \left\{ 0, 1 \right\}$ with $\phi(x) = 1 \iff x \not \in \mathcal{N}$. ...

Active Learning concerns methods to decide how to sample the most useful information in a specific domain; how can you select the best sample for an unknown model? Gathering data is very costly, we would like to create some principled manner to choose the best data point to humanly label in order to have the best model. In this setting, we are interested in the concept of usefulness of information. One of our main goals is to reduce uncertainty, thus, Entropy-based (mutual information) methods are often used. For example, we can use active learning to choose what samples needs to be labelled in order to have highest accuracy on the trained model, when labelling is costly. ...

Bayesian Information Criterion (BIC) The Bayesian Information Criterion (BIC) is a model selection criterion that helps compare different statistical models while penalizing model complexity. It is rooted in Bayesian probability theory but is commonly used even in frequentist settings. Mathematically Precise Definition For a statistical model $M$ with $k$ parameters fitted to a dataset $\mathcal{D} = \{x_1, x_2, \dots, x_n\}$, the BIC is defined as: $$ \text{BIC} = -2 \cdot \ln \hat{L} + k \cdot \ln(n) $$where: ...

The beta distribution The beta distribution is a powerful tool for modeling probabilities and proportions between 0 and 1. Here’s a structured intuition to grasp its essence: Core Concept The beta distribution, defined on $[0, 1]$, is parameterized by two shape parameters: α (alpha) and β (beta). These parameters dictate the distribution’s shape, allowing it to flexibly represent beliefs about probabilities, rates, or proportions. Key Intuitions a. “Pseudo-Counts” Interpretation α acts like “successes” and β like “failures” in a hypothetical experiment. Example: If you use Beta(5, 3), it’s as if you’ve observed 5 successes and 3 failures before seeing actual data. After observing x real successes and y real failures, the posterior becomes Beta(α+x, β+y). This makes beta the conjugate prior for the binomial distribution (bernoulli process). b. Shape Flexibility Uniform distribution: When α = β = 1, all values in [0, 1] are equally likely. Bell-shaped: When α, β > 1, the distribution peaks at mode = (α-1)/(α+β-2). Symmetric if α = β (e.g., Beta(5, 5) is centered at 0.5). U-shaped: When α, β < 1, density spikes at 0 and 1 (useful for modeling polarization, meaning we believe the model to only produce values at 0 or 1, not in the middle.). Skewed: If α > β, skewed toward 1; if β > α, skewed toward 0. c. Moments Mean: $α/(α+β)$ – your “expected” probability of success. Variance: $αβ / [(α+β)²(α+β+1)]$ – decreases as α and β grow (more confidence). $$ \text{Mode} = \frac{\alpha - 1}{\alpha + \beta - 2} $$The mathematical model $$ \text{Beta} (x \mid a, b) = \frac{1}{B(a, b)} \cdot x^{a -1 }(1 - x)^{b - 1} $$ Where $B(a, b) = \Gamma(a) \Gamma(b) / \Gamma( + b)$ And $\Gamma(t) = \int_{0}^{\infty}e^{-x}x^{t - 1} \, dx$ ...

Machine learning cannot distinguish between causal and environment features. Shortcut learning Often we observe shortcut learning: the model learns some dataset dependent shortcuts (e.g. the machine that was used to take the X-ray) to make inference, but this is very brittle, and is not usually able to generalize. Shortcut learning happens when there are correlations in the test set between causal and non-causal features. Our object of interest should be the main focus, not the environment around, in most of the cases. For example, a camel in a grass land should still be recognized as a camel, not a cow. One solution could be engineering invariant representations which are independent of the environment. So having a kind of encoder that creates these representations. ...

There is a big difference between the empirical score and the expected score; in the beginning, we had said something about this in Introduction to Advanced Machine Learning. We will develop more methods to better comprehend this fundamental principles. How can we estimate the expected risk of a particular estimator or algorithm? We can use the cross-validation method. This method is used to estimate the expected risk of a model, and it is a fundamental method in machine learning. ...

The DP (Dirichlet Processes) is part of family of models called non-parametric models. Non parametric models concern learning models with potentially infinite number of parameters. One of the classical application is unsupervised techniques like clustering. Intuitively, clustering concerns in finding compact subsets of data, i.e. finding groups of points in the space that are particularly close by some measure. The Dirichlet Process See Beta and Dirichlet Distributions for the definition and intuition of these two distributions. One quite important thing that Dirichlet allows to do is the ability of assigning an ever growing number of clusters to data. This models are thus quite flexible to change and growth. ...

The idea of ensemble methods goes back to Sir Francis Galton. In 787, he noted that although not every single person got the right value, the average estimate of a crowd of people predicted quite well. The main idea of ensemble methods is to combine relatively weak classifiers into a highly accurate predictor. The motivation for boosting was a procedure that combines the outputs of many “weak” classifiers to produce a powerful “committee.” ...

A simple motivation Fisher’s Linear Discriminant is a simple idea used to linearly classify our data. The image above, taken from (Bishop 2006), is the summary of the idea. We clearly see that if we first project using the direction of maximum variance (See Principal Component Analysis) then the data is not linearly separable, but if we take other notions into consideration, then the idea becomes much more cleaner. ...