Fernando Silveira

FERNANDO "KAVINSKY" SILVEIRA

AI Researcher · Neuroevolution · XAI · Visual Analytics

UFRGS · INF · ExpLAIn

ROBCO INDUSTRIES (TM) TERMINAL v3.1.2

[ > ] ABOUT

I am an engineer and M.Sc. researcher in Artificial Intelligence at the Federal University of Rio Grande do Sul (UFRGS). My work focuses on neuroevolution, population-based learning, and the visualization of learning dynamics in neural network weight spaces.

My background spans software engineering, aeronautical sciences, and machine learning, with emphasis on interpretable, reliable, and reproducible AI systems for scientific and safety-critical domains.

[ > ] CURRENTLY INVESTIGATING

Understanding how neural populations evolve through weight space.

  • Neuroevolutionary population dynamics
  • Population trajectory alignment in weight space
  • Weight-space manifold visualization
  • Interpretable evolutionary systems

[ > ] EDUCATION

M.Sc. in Artificial Intelligence — UFRGS

Research on neuroevolutionary population dynamics in neural network weight spaces using dimensionality reduction, temporal alignment, and vector field analysis.

B.Sc. in Software Engineering

Software systems design, distributed systems, and engineering methodologies.

B.Sc. in Aeronautical Sciences

Aviation systems, operational safety, and aeronautical analysis.

B.Sc. in Systems Analysis and Development

Software development and information systems.

Postgraduate Studies

Data Science · Business Intelligence · Process Management · Project Management

Education

[ > ] RESEARCH INTERESTS

Research Interests

My research explores neuroevolution, evolutionary computation, and the dynamics of learning in neural network parameter spaces. I am particularly interested in visualization methods for population-based learning systems, interpretable AI, and reliable machine learning for scientific and safety-critical applications.

Core Topics

Neuroevolution · Evolutionary Computation · Machine Learning · Representation Dynamics

Explainability & Visualization

XAI · Weight Space Analysis · Dimensionality Reduction · Visual Analytics

Applied Domains

Aviation Systems · Bioinformatics · Scientific Computing

Engineering

Research Software · Reproducibility · ML Infrastructure

[ > ] TECH STACK

Languages & Frameworks

Python · PyTorch · JAX · TensorFlow · CUDA

Analysis & ML

UMAP · t-SNE · Scikit-learn · OpenCV · Pandas

Infrastructure

Docker · Linux · FastAPI · PostgreSQL · Git

Visualization

Matplotlib · Plotly · Streamlit · D3.js

[ > ] AERONAUTICS

Beyond artificial intelligence research, I maintain a strong interest in aeronautics, aviation systems, and aerospace engineering.

I am particularly interested in the intersection of AI and aviation, including autonomous systems, trajectory optimization, anomaly detection, and safety-critical decision support.

Aviation provides a unique environment for developing interpretable and reliable machine learning systems due to its operational complexity and strict safety requirements.

I am currently pursuing Private Pilot training while studying Aeronautical Sciences, connecting practical aviation experience with AI research and engineering._

Aeronautics

[ > ] PROJECTS

CuMiDaVis Dashboard

CuMiDaVis

Visual analytics dashboard for cancer microarray datasets using PCA, t-SNE, and UMAP for high-dimensional representation analysis.

GeneCancerPredictor

GeneCancerPredictor

Machine learning framework for hepatocellular carcinoma prediction comparing six classification models with hyperparameter optimization and evaluation pipelines.

neuroevo-weight-space-viz

neuroevo-weight-space-viz

Visualization framework for neuroevolutionary populations in neural network weight spaces using aligned UMAP embeddings, vector fields, and trajectory analysis.

[ > ] PUBLICATIONS

Publications and research manuscripts coming soon.

Current research topics:

  • Neuroevolutionary population dynamics
  • Weight space visualization
  • Evolutionary representation analysis
  • Explainable AI systems

[ > ] LABORATORY

ExpLAIn Lab

ExpLAIn – Explainability Laboratory for Artificial Intelligence

Instituto de Informática – UFRGS, Porto Alegre, Brasil

Research group focused on interpretable, reliable, and responsible artificial intelligence systems.

  • Interpretable Machine Learning
  • AI for Biology & Biomarker Discovery
  • Evolutionary Optimization of Neural Networks
  • Visual Analytics for AI Systems

Coordination: Prof. Bruno Iochins Grisci (PhD in Computer Science)

[ > ] CONTACT

Open to research collaborations, academic partnerships, and opportunities involving artificial intelligence, neuroevolution, visual analytics, and aviation systems.

Contact