Advanced Topics in Machine Learning Caltech, Spring 2023

Additional references will be added here soon. Stay tuned!

Gaussian Processes

• Scalable Variational Gaussian Process Classification

• GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration (GPyTorch website)

• Exact Gaussian Processes on a Million Data Points

• Scalable Cross Validation Losses for Gaussian Process Models

• Variational Nearest Neighbor Gaussian Process

• Sharp Calibrated Gaussian Processes

Bayesian Deep Learning and Approximate Bayesian Inference

• Auto-Encoding Variational Bayes

• Bayesian Neural Networks Tutorial

• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning

• Weight Uncertainty in Neural Networks

• Bayesian Deep Learning and a Probabilistic Perspective of Generalization

• Accurate Uncertainties for Deep Learning Using Calibrated Regression

• Evaluating Approximate Inference in Bayesian Deep Learning

• Randomized Prior Functions for Deep Reinforcement Learning

Adaptive Experimentation

• A Tutorial on Bayesian Optimization

• Thinking Inside the Box: A tutorial on Grey-Box Bayesian optimization

• Scalable Global Optimization via Local Bayesian Optimization

• BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization (BoTorch website)

• Bayesian Optimization over Discrete and MixeSpaces via Probabilistic Reparameterization

• Neural Thompson Sampling

• Deep Kernel Bayesian Optimization

• Local Latent Space Bayesian Optimization over Structured Inputs

• Bayesian Optimal Active Search and Surveying

• The K-armed Dueling Bandit Problem

• qEUBO: A Decision-Theoretic Acquisition Function for Preferential Bayesian Optimization

• Efficient Nonmyopic Bayesian Optimization via One-Shot Multi-Step Trees

• Bayesian Optimization of Function Networks

• Causal-BALD: Deep Bayesian Active Learning of Outcomes to Infer Treatment-Effects from Observational Data

• Targeted Causal Elicitation

• Differentiable Multi-Target Causal Bayesian Experimental Design

Conformal Prediction

• A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification

• Conformal Prediction Beyond Exchangeability

Applications

• Deep Probabilistic Imaging: Uncertainty Quantification and Multi-modal Solution Characterization for Computational Imaging

• Efficient Uncertainty Quantification from Exoplanet Astrometry to Black Hole Feature Extraction

• Multi-information Source Bayesian Optimization of Culture Media for Cellular Agriculture

• Constrained Bayesian Optimization for Automatic Chemical Design Using Variational Autoencoders

• Efficient Nonmyopic Active Search with Applications in Drug and Materials Discovery

• Bayesian Uncertainty Estimation of Learned Variational MRI Reconstruction

• Preference-Based Learning for Exoskeleton Gait Optimization