Machine Learning Engineer
CurrentMicrosoft Professional Program for Artificial Intelligence - https://academy.microsoft.com/en-us/tracks/artificial-intelligence (140 hours lecture + hands-on training - see certifications for individual courses)• Python (w/Numpy and Pandas), CNTK, Azure Machine Learning• Ethics and Law in Data and Analytics• Creating data studies, making valid claims and statistical rigor. Data visualization and cleaning.• Supervised and unsupervised standard machine learning algorithms• Multi-Layered Perceptron (MLP), Convolution Neural Networks (CNNs), Recurrent Neural Networks (RNNs) and Long Short Term Memory Networks (LSTMs)• Reinforcement Learning - policy creation and optimization with Dynamic Programming (DP), Temporal Difference Learning (TD), Actor-Critic, Deep-Q learning• Natural Language Processing (NLP) - Character/N-Gram representations, Deep Semantic Similarity Models (DSSMs), Deep Reinforcement Learning for NLP, Machine Translation, Image Captioning• Capstone Project - Analyze current and voltage spectrogram images (including startup and continuous operation images) to determine type of appliance being measured (11 unit classification problem.) Solution using multi-layered CNN, with fully connected final layer.Deeplearning.ai (Andrew Ng) - https://www.deeplearning.ai/deep-learning-specialization (45 hours of lecture + hands on training - see Certifications for individual courses)• Python, Tensorflow, Keras• MLPs, CNNs, RNNs, LSTMs, Gated Recurrent Units (GRUs), ResNets• NLP - word embeddings, machine translation and image captioning• Computer Vision - Object detection and Face Recognition• Regularization (L2, L1, Dropout, Pooling)• Hyperparameter Tuning Strategies, Error Analysis and Model Iteration strategies• Transfer Learning Opportunities, Best Practices and Available assets• Managing Deep Learning Projects• Mathematics for Vectorization and Back-propagation
