Paulito Palmes, PhD

I am a research scientist at the IBM Research working on the following areas: AutoML, AutoAI, RL/ML Optimization, and Decision Optimization.


Finding an Effective Strategy for AutoML Pipeline Optimization
Paulito Palmes, PhD

One of the main problems in AutoML implementation is finding the best strategy to search the most optimal pipeline in prediction or classification tasks. This problem is commonly known as CASH (Combined Algorithm Selection and Hyperparameter Optimization). This talk will show competitive results with significantly shorter computation time by just focusing the search in the model selection and structure of the pipeline without the need of hyperparameter optimization.

Lale in Julia: A package for semi-automated data science
Paulito Palmes, PhD, Kiran Kate

Lale.jl is a Julia wrapper of lale Python library for semi-automated data science. Lale.jl offers scikit-learn compatible AutoML with algorithm selection and hyper-parameter tuning. Lale.jl provides a consistent high-level interface to existing AutoML optimizer backends such as Hyperopt, GridSearchCV, and SMAC. It has a standard search space specification with out of the box search space for 180 operators from scikit-learn, imblearn, AIF360, SnapML and more.

Data driven insight into fish behaviour for aquaculture
Paulito Palmes, PhD, Fearghal O'Donncha

Aquaculture, or the farmed production of fish and shellfish, has grown rapidly, from supplying just 7% of fish for human consumption in 1974 to more than half in 2016. Sustaining this rapid expansion requires data-driven management of the production process and environmental impacts. This talk presents a machine-learning-based exploration of environmental and fish behaviour datasets collected at three salmon farms in Norway, Scotland, and Canada using AutoML tools in Julia.