The document discusses automated machine learning (AutoML). It defines AutoML as providing methods to make machine learning more efficient and accessible to non-machine learning experts. AutoML aims to automate tasks like data preprocessing, feature engineering, algorithm selection and hyperparameter optimization. This can reduce costs, increase productivity for data scientists and democratize machine learning. The document also lists several AutoML tools that provide hyperparameter tuning, full pipeline optimization or neural architecture search.
2. WHAT IS MACHINE LEARNING ?
Machine Learning, at its most basic form, is the practice of using
algorithms to parse data, learn from it, and then make a
determination or prediction about something in the world.
3. MACHINE LEARNING
TECHNIQUES
• Supervised Learning
• Classification
• Regression
• Unsupervised Learning
• Clustering
• Association
• Deep Learning
• CNN
• ANN
• Reinforcement Learning
8. WHAT IS AUT. ML. ?
• Automated Machine Learning provides methods and processes to make Machine
Learning available for non-Machine Learning experts, to improve efficiency of
Machine Learning and to accelerate research on Machine Learning.
9. AUTOMATED MACHINE LEARNING INCLUDES
Automated ML solutions aim to automate some or all steps of the machine learning
process which includes:
• Data pre-processing
• Feature engineering
• Feature extraction
• Feature selection
• Algorithm selection & hyperparameter optimization
10. WHAT IS BENEFIT OF AUT. ML. ?
• Cost reductions
• Increased productivity for data scientists
• Democratization of machine learning reduces demand for data scientists
• Increased revenues and customer satisfaction
• Rolling out more models with increased accuracy improves business results
12. WHY IS AUT. ML. IS IMPORTANT?
• Manually constructing a machine learning model is a multistep process that
requires domain knowledge, mathematical expertise, and computer science skills –
which is a lot to ask of one company, let alone one data scientist (provided you can
hire and retain one)
13. WHY IS AUT. ML. IS IMPORTANT?
• Automated machine learning enables organizations to use the baked-in knowledge
of data scientists without having to develop the capabilities themselves,
simultaneously improving return on investment in data science initiatives and
reducing the amount of time it takes to capture value
14. TOOLS
Hyperparameter optimization and model selection
• H2O AutoML
• provides automated data preparation, hyperparameter tuning via random search, and
stacked ensembles in a distributed machine learning platform.)
• Mir
• a R package that contains several hyperparameter optimization techniques for machine
learning problems.
15. TOOLS
Hyperparameter optimization and model selection
• In machine learning, hyperparameter optimization ortuning is the problem of
choosing a set of optimalhyperparameters for a learning algorithm. ... These
measures are called hyperparameters, and have to be tuned so that the model can
optimally solve the machine learning problem.
16. TOOLS
Full Pipeline Optimization
• The goal of this ML pipeline is to gather data on inventory, users, and advertiser
information, and to train ML models to predict the likelihood of someone clicking on
an ad, at the time of auction. This is one of the core optimization parts of the Ad
Tech business.
17. TOOLS
Full Pipeline Optimization
• AutoWeka
• is a Bayesian hyperparameter optimization layer on top of WEKA.
• Auto-Sklearn
• is a Bayesian hyperparameter optimization layer on top of scikit-learn.
• Firefly.ai
• a Cloud-Based system for automatic generation of machine learning models
• TPOT
• is a Python library that automatically creates and optimizes full machine learning pipelines using genetic
programming.
• TransmogrifAI
• is a Scala/SparkML library created by Salesforce for automated data cleansing, feature engineering, model
selection, and hyperparameter optimization
• RECIPE
• is a framework based on grammar-based genetic programming that builds customized scikit-learn
classification pipelines.
18. TOOLS
Deep neural network search
• Neural architecture search (NAS) uses machine learning to automate the design
of artificial neural networks. Various approaches to NAS have designed networks
that compare well with hand-designed systems. The basic search algorithm is to
propose a candidate model, evaluate it against a dataset and use the results as
feedback to teach the NAS network
19. TOOLS
Deep neural network search
• Devol
• is a Python package that performs Deep Neural Network architecture search
using genetic programming.
• Google AutoML
• for deep learning model architecture selection.
• Auto Keras
• is an open-source python package for neural architecture search.