Andrew Silva
Machine Learning Mentor @The Erdős Institute
Signup · Get unlimited contacts
WORK HISTORY
Machine Learning Mentor @The Erdős Institute
Machine learning models, including-Classification (logistic regression, LDA, QDA, Naive Bayes, KNN)-Ensemble methods (regularization, Lasso, Ridge, ElasticNet, bagging, boosting)-Dimensionality reduction (PCA, PLS)-Decision trees, random forests-Neural networks (CNN, LSTM, RNN)-Cross validation, bootstrapping-Unsupervised learning (k-means, hierarchical clustering)NLP neural network model using text reviews to predict star rating classification-Data from Google Local entity reviews-Logistic model using bag-of-words encoding-Bi-directional LSTM model, custom embedding layer-Bi-directional LSTM model, GloVe 6b embedding layer-BERT, with pre-trained frozen and unfrozen weights-RoBERTa, pre-trained weights on review sentiment-Best model attained 92% accuracy for 5-level classification using 1% of available data
EDUCATION
University of California, Davis
B.S.
National University of Singapore
Visiting Researcher
Geneva Graduate Institute
Ph.D.
University of California, Davis
B.A.
Graduate Institute of International and Development Studies
M.Sc.
ABOUT ANDREW SILVA
Data Scientist and PhD Economist with 12+ years experience in causal inference statistics…
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.