Thread by @mparmar47

[1/7] #AutoML (Automated Machine learning) is a new buzzword in the evolution of AI and ML. AutoML is the process of automating the end-to-end process of applying #machinelearning to real-world problems.
#AI #likeabosch #innovation #algorithms #machinelearning

[2/7] I recently read an interesting paper on the topic of #AutoML. form it, I created a taxonomy of automl parameters as a quick summary of papers, and at the top level it has three parts. #AI #likeabosch #innovation #algorithms #machinelearning

[3/7] Metalearning: Process of learning from previous experience gained during applying various learning algorithms on different kinds of data, & hence reducing the needed time to learn new tasks for a model. #AI #likeabosch #algorithms #machinelearning #innovation #AutoML

[4/7] Neural Architecture Search( #NAS): It is an automated process to design the architecture of #ANN (Artificial neural networks) for #DeepLearning models. #AI #likeabosch #innovation #algorithms #machinelearning #AutoML

[5/7] HyperParameter Optimisation: It is a process of tuning the hyper-parameters of the model to further optimize the model performance #AI #likeabosch #innovation #algorithms #machinelearning #innovation #AutoML

[6/7] If you are further interested to know in detail about #AUtoML, refer the original paper "Automated Machine Learning: State-of-The-Art and Open Challenges" arxiv.org/abs/1906.02287

[7/7] The picture depicts a detailed breakdown of taxonomy of #AutoML parameters
#AI #likeabosch #innovation #algorithms #machinelearning #innovation