Journal

Reinforcement Learning(RL) trains an agent to take actions in an environment, the agent does not get told what to action correct, it basically gets a reward signal instead.

Principal component analysis(PCA) allows us to take data with a lot of features say hundreds or thousands of features and reduce it to very small number of features say two or five features.

Recommender system predicts how a user would rank or respond to items they haven't seen yet, then surface with the highest predicted score

Anomaly detection looks at unlabeled dataset of normal examples, learns what normal looks like, then flags new examples that looks different

Clustering looks at a dataset and tries to find points that are similar to each other and group them together

Learned about the Gini index as a measure of impurity for decision trees

Learned a whole bunch about decision trees.

Learned about model diagnosis, when a model performs poorly, we can do the following: - Collect more training data - Try a smaller set of features - Add more features - Add polynomial features - Decrease the regularization parameter $\lambda$ - Increase $\lambda$ - Make the model bigger - Train longer

Learned about the steps involved in training a neural network which are: specify the model, specify the loss and cost function and then train the model on the training dataset to minimize the cost function.

Neuron is a computational unit which takes a vector of input, computes a weighted sum plus a bias, then apply a non-linear activation function to produce a single output.

Learned classification with logistic regression

Learned about regression with multiple input variables

No ML today

Not much

Explored intro to machine learning and types of machine learning models

Learned about Confidence Intervals and Margin of Error

Learned about about Sampling, Law of Large Numbers and Central Limit Theorem

Learned about expected value, variance, standard deviation, skewness and kurtosis

Learned about use or probability and statistics in ML

Learned how to use calculus for back propagation in training a neural network with simple layers(input and output, no hidden layer) and multiple perceptron for both linear regression and classification problems

Learned about use of Calculus in ML

Learned about Variance and Covariance and how they related to spread in a dataset

Learned about Variance and Covariance and how they related to spread in a dataset

Learned about Vectors and Linear Transformations

Learning about Mathplotlib for data visualization

Learned a whole bunch of things(most things I've done before), learn about solving systems of linear equations, rank of a matrix, echelon form, diagonal matrix and Gaussian elimination method

Build something for the Zama builder bounty

Learned about python basics such as variables, lists and loops

PyTorch is the dominant framework in research and modern ML — TensorFlow has lost the research community