Portfolio

Scientific Machine Learning with Symbolic Regression

Published:

In this project, our objective is to develop a Symbolic Regression algorithm capable of identifying symbolic expressions while adhering to predefined functional forms and shape constraints. These constraints are carefully chosen to align with the specific needs and expectations of the collected data. The endeavor necessitates an interdisciplinary approach, drawing from various scientific disciplines to comprehend the diverse desiderata and constraints at play. We will apply advanced regression analysis concepts to achieve this goal.

Functional Program Synthesis

Published:

In this project we aim at creating an algorithm that generates programs in Haskell, a pure functional programming language, exploiting many useful programming patterns and the type leve information extracted from the program specification.

Scientific Computing in Haskell

Published:

In this project, we will build a pure Haskell library using the array library called Massiv that implements the main scientific computing algorithms with a comparable performance to these well known libraries (a SciHask).

Automatic Parallelism for Stencils using Comonads

Published:

Neste projeto de pesquisa será feita a integração de técnicas de paralelismo para stencils com o conceito de comônadas, típico de linguagens funcionais. O objetivo é permitir que um programador de uma linguagem funcional possa (de uma maneira simples, eficiente e paralela) definir, executar e obter o resultado da execução de um stencil.

Statistical Tools for Symbolic Regression

Published:

In the scope of this project, our aim is to implement and adapt such statistical tools, with a specific focus on extending the capabilities of the srtree-opt program. This program is capable of parsing and processing a multitude of symbolic regression models. Through these adaptations and enhancements, we intend to bridge the gap between Symbolic Regression and the extensive statistical toolkit available for traditional regression analysis, ultimately elevating the analytical capabilities in this domain.

Type-Safe Metaheuristics

Published:

The primary goal of this project is to develop a type-safe domain-specific language (DSL) in Haskell. This DSL will facilitate the description of key metaheuristic algorithms, enabling end users to effortlessly experiment with various combinations of available search operators. It goes beyond merely implementing vanilla versions of these algorithms; it empowers users to explore hybrid approaches as well. Furthermore, this tool will incorporate a native concurrency module, allowing users to harness the full potential of multicore machines for enhanced efficiency and performance.