ARTIFICIAL INTELLIGENCE MASTERY CERTIFICATION GADET201
Profiles that can prepare this certification contents:
Data engineer, Data scientist, Developer engineer.
Please contact us at firstname.lastname@example.org to receive the materials and the virtual machine to prepare this certification.
Global knowledge to be acquired to pass this certification:
- Artificial intelligence basics
- Supervised learning regression
- Supervised learning classification
- Unsupervised learning
- Deep Learning
Detailed plan of preparation:
Artificial Intelligence basics
Supervised learning regression
Simple and multiple linear regression. Polynomial regression. Evaluation metrics. Mean Squared Error. Absolute Squared Error. Root Mean Squared Error. R2. Adjusted R2. Features selection. P values. Features importance. Sci-Kit Learn library.
Supervised learning classification
Nearest Neighbors. Logistic regression. Support Vector Machine (linear and kernel). Decision Tree. Random Forest. Naïve Bayes. Artificial Neural Networks. Evaluation metrics. Accuracy. Precision. Recall. F2 score. ROC curve. CAP curve. Sci-Kit Learn library. Tensoflow and Keras libraries.
K-Means. Hierarchical Clustering. Dimensionality reduction with Principal Component. Clusters visualization. Sci-Kit Learn library.
deep feed-forward neural network
(a) Machine learning in the supervised context
(b) The formal neuron
(c) Neural networks
(d) Gradient descent algorithm
(e) Deep Learning: CPU, GPU and cluster
(f) Theano, Tensorflow and Keras
(g) Deep feed-forward neural network
Convolutional and recurrent neural network
1) Convolutional Neural Networks
(a) Convolution properties
(b) Convolution layers
(c) Consolidation (pooling)
2) Recurrent Neural Network (Recurrent Neural Networks)
(a) Prediction of time series
(b) LSTM Recurrent Neural Networks
(c) Short-term and long-term memory recurrent neural network for regression (LSTM Network For Regression)
(d) Short-term and long-term memory recurrent neural network using the window method (LSTM For Regression Using the Window Method)
(e) Short-term and long-term memory recurrent neural network with time steps (LSTM For Regression with Time Steps)