Skip to content

blmoistawinde/ml_equations_latex

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Classical ML Equations in LaTeX

A collection of classical ML equations in Latex . Some of them are provided with simple notes and paper link. Hopes to help writings such as papers and blogs.

Better viewed at https://blmoistawinde.github.io/ml_equations_latex/

Model

RNNs(LSTM, GRU)

encoder hidden state math at time step math , with input token embedding math

math

decoder hidden state math at time step math , with input token embedding math

math

h_t = RNN_{enc}(x_t, h_{t-1})
s_t = RNN_{dec}(y_t, s_{t-1})

The math , math are usually either

Attentional Seq2seq

The attention weight math , the math th decoder step over the math th encoder step, resulting in context vector math

math

math

math

c_i = \sum_{j=1}^{T_x} \alpha_{ij}h_j

\alpha_{ij} = \frac{\exp(e_{ij})}{\sum_{k=1}^{T_x} \exp(e_{ik})}

e_{ij} = a(s_{i-1}, h_j)

math is an specific attention function, which can be

Bahdanau Attention

Paper: Neural Machine Translation by Jointly Learning to Align and Translate

math

e_{ij} = v^T tanh(W[s_{i-1}; h_j])

Luong(Dot-Product) Attention

Paper: Effective Approaches to Attention-based Neural Machine Translation

If math and math has same number of dimension.

math

otherwise

math

e_{ij} = s_{i-1}^T h_j

e_{ij} = s_{i-1}^T W h_j

Finally, the output math is produced by:

math

math

s_t = tanh(W[s_{t-1};y_t;c_t])
o_t = softmax(Vs_t)

Transformer

Paper: Attention Is All You Need

Scaled Dot-Product attention

math

Attention(Q, K, V) = softmax(\frac{QK^T}{\sqrt{d_k}})V

where math is the dimension of the key vector math and query vector math .

Multi-head attention

math

where

math

MultiHead(Q, K, V) = Concat(head_1, ..., head_h)W^O

head_i = Attention(Q W^Q_i, K W^K_i, V W^V_i)

Generative Adversarial Networks(GAN)

Paper: Generative Adversarial Networks

Minmax game objective

$$\min_{G}\max_{D}V(D, G) = \mathbb{E}_{x\sim p_{\text{data}}(x)}[\log{D(x)}] + \mathbb{E}_{z\sim p_{\text{z}}(z)}[\log{(1-D(G(z)))}]$$
\min_{G}\max_{D}V(D, G) = \mathbb{E}_{x\sim p_{\text{data}}(x)}[\log{D(x)}] +  \mathbb{E}_{z\sim p_{\text{z}}(z)}[\log{(1-D(G(z)))}]

Variational Auto-Encoder(VAE)

Paper: Auto-Encoding Variational Bayes

Reparameterization trick

To produce a latent variable z such that math , we sample math , than z is produced by

math

z \sim q_{\mu, \sigma}(z) = \mathcal{N}(\mu, \sigma^2)
\epsilon \sim \mathcal{N}(0,1)
z = \mu + \epsilon \cdot \sigma

Above is for 1-D case. For a multi-dimensional (vector) case we use:

math

math

\epsilon \sim \mathcal{N}(0, \textbf{I})
\vec{z} \sim \mathcal{N}(\vec{\mu}, \sigma^2 \textbf{I})

Activations

Sigmoid

Related to Logistic Regression. For single-label/multi-label binary classification.

math

\sigma(z) = \frac{1} {1 + e^{-z}}

Tanh

math

tanh(x) = \frac{e^x - e^{-x}}{e^x + e^{-x}} = \frac{1 - e^{-2x}}{1 + e^{-2x}}

Softmax

For multi-class single label classification.

math

\sigma(z_i) = \frac{e^{z_{i}}}{\sum_{j=1}^K e^{z_{j}}} \ \ \ for\ i=1,2,\dots,K

Relu

math

Relu(z) = max(0, z)

Gelu

math

where math is the cumulative distribution function of Gaussian distribution.

Gelu(x) = x\Phi(x)

Loss

Regression

Below math and math are math dimensional vectors, and math denotes the value on the math th dimension of math .

Mean Absolute Error(MAE)

math

\sum_{i=1}^{D}|x_i-y_i|

Mean Squared Error(MSE)

math

\sum_{i=1}^{D}(x_i-y_i)^2

Huber loss

It’s less sensitive to outliers than the MSE as it treats error as square only inside an interval.

math

L_{\delta}=
    \left\{\begin{matrix}
        \frac{1}{2}(y - \hat{y})^{2} & if \left | (y - \hat{y})  \right | < \delta\\
        \delta ((y - \hat{y}) - \frac1 2 \delta) & otherwise
    \end{matrix}\right.

Classification

Cross Entropy

  • In binary classification, where the number of classes math equals 2, Binary Cross-Entropy(BCE) can be calculated as:

math

  • If math (i.e. multiclass classification), we calculate a separate loss for each class label per observation and sum the result.

math

-{(y\log(p) + (1 - y)\log(1 - p))}

-\sum_{c=1}^My_{o,c}\log(p_{o,c})

M - number of classes

log - the natural log

y - binary indicator (0 or 1) if class label c is the correct classification for observation o

p - predicted probability observation o is of class c

Negative Loglikelihood

math

Minimizing negative loglikelihood

math

is equivalent to Maximum Likelihood Estimation(MLE).

math

Here math is a scaler instead of vector. It is the value of the single dimension where the ground truth math lies. It is thus equivalent to cross entropy (See wiki).\

NLL(y) = -{\log(p(y))}

\min_{\theta} \sum_y {-\log(p(y;\theta))}

\max_{\theta} \prod_y p(y;\theta)

Hinge loss

Used in Support Vector Machine(SVM).

math

max(0, 1 - y \cdot \hat{y})

KL/JS divergence

math

math

KL(\hat{y} || y) = \sum_{c=1}^{M}\hat{y}_c \log{\frac{\hat{y}_c}{y_c}}

JS(\hat{y} || y) = \frac{1}{2}(KL(y||\frac{y+\hat{y}}{2}) + KL(\hat{y}||\frac{y+\hat{y}}{2}))

Regularization

The math below can be any of the above loss.

L1 regularization

A regression model that uses L1 regularization technique is called Lasso Regression.

math

Loss = Error(Y - \widehat{Y}) + \lambda \sum_1^n |w_i|

L2 regularization

A regression model that uses L1 regularization technique is called Ridge Regression.

math

Loss = Error(Y - \widehat{Y}) +  \lambda \sum_1^n w_i^{2}

Metrics

Some of them overlaps with loss, like MAE, KL-divergence.

Classification

Accuracy, Precision, Recall, F1

math

math

math

math

Accuracy = \frac{TP+TN}{TP+TN+FP+FN}
Precision = \frac{TP}{TP+FP}
Recall = \frac{TP}{TP+FN}
F1 = \frac{2*Precision*Recall}{Precision+Recall} = \frac{2*TP}{2*TP+FP+FN}

Sensitivity, Specificity and AUC

math

math

Sensitivity = Recall = \frac{TP}{TP+FN}
Specificity = \frac{TN}{FP+TN}

AUC is calculated as the Area Under the math (TPR)- math (FPR) Curve.

Regression

MAE, MSE, equation above.

Clustering

(Normalized) Mutual Information (NMI)

The Mutual Information is a measure of the similarity between two labels of the same data. Where math is the number of the samples in cluster math and math is the number of the samples in cluster math , the Mutual Information between cluster math and math is given as:

math

MI(U,V)=\sum_{i=1}^{|U|} \sum_{j=1}^{|V|} \frac{|U_i\cap V_j|}{N}
\log\frac{N|U_i \cap V_j|}{|U_i||V_j|}

Normalized Mutual Information (NMI) is a normalization of the Mutual Information (MI) score to scale the results between 0 (no mutual information) and 1 (perfect correlation). In this function, mutual information is normalized by some generalized mean of H(labels_true) and H(labels_pred)), See wiki.

Skip RI, ARI for complexity.

Also skip metrics for related tasks (e.g. modularity for community detection[graph clustering], coherence score for topic modeling[soft clustering]).

Ranking

Skip nDCG (Normalized Discounted Cumulative Gain) for its complexity.

(Mean) Average Precision(MAP)

Average Precision is calculated as:

math

\text{AP} = \sum_n (R_n - R_{n-1}) P_n

where math and math are the precision and recall at the math th threshold.

AP can also be regarded as the area under the precision-recall curve.

MAP is the mean of AP over all the queries.

Similarity/Relevance

Cosine

math

Cosine(x,y) = \frac{x \cdot y}{|x||y|}

Jaccard

Similarity of two sets math and math .

math

Jaccard(U,V) = \frac{|U \cap V|}{|U \cup V|}

Pointwise Mutual Information(PMI)

Relevance of two events math and math .

math

PMI(x;y) = \log{\frac{p(x,y)}{p(x)p(y)}}

For example, math and math is the frequency of word math and math appearing in corpus and math is the frequency of the co-occurrence of the two.

Notes

This repository now only contains simple equations for ML. They are mainly about deep learning and NLP now due to personal research interests.

For time issues, elegant equations in traditional ML approaches like SVM, SVD, PCA, LDA are not included yet.

Moreover, there is a trend towards more complex metrics, which have to be calculated with complicated program (e.g. BLEU, ROUGE, METEOR), iterative algorithms (e.g. PageRank), optimization (e.g. Earth Mover Distance), or even learning based (e.g. BERTScore). They thus cannot be described using simple equations.

Reference

Pytorch Documentation

Scikit-learn Documentation

Machine Learning Glossary

Wikipedia

https://blog.floydhub.com/gans-story-so-far/

https://ermongroup.github.io/cs228-notes/extras/vae/

Thanks for a-rodin's solution to show Latex in Github markdown, which I have wrapped into latex2pic.py.

About

Classical ML equations in Latex, helps paper and blog writing

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published