Model Inversion Networks for Model-Based Optimization

Abstract

In this work, we aim to solve data-driven optimization problems, where the goal is to find an input that maximizes an unknown score function given access to a dataset of inputs with corresponding scores. When the inputs are high-dimensional and valid inputs constitute a small subset of this space (e.g., valid protein sequences or valid natural images), such model-based optimization problems become exceptionally difficult, since the optimizer must avoid out-of-distribution and invalid inputs. We propose to address such problem with model inversion networks (MINs), which learn an inverse mapping from scores to inputs. MINs can scale to high-dimensional input spaces and leverage offline logged data for both contextual and non-contextual optimization problems. MINs can also handle both purely offline data sources and active data collection. We evaluate MINs on tasks from the Bayesian optimization literature, high-dimensional model-based optimization problems over images and protein designs, and contextual bandit optimization from logged data.

References in corpus (9)

• Auto-Encoding Variational Bayes
• Conditional Generative Adversarial Nets
• Practical Bayesian Optimization of Machine Learning Algorithms
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Categorical Reparameterization with Gumbel-Softmax
• Pixel Recurrent Neural Networks
• Deep reinforcement learning from human preferences
• Counterfactual Risk Minimization: Learning from Logged Bandit Feedback
• Feedback GAN (FBGAN) for DNA: a Novel Feedback-Loop Architecture for Optimizing Protein Functions

Cited by in corpus (7)

• Plug & Play Directed Evolution of Proteins with Gradient-based Discrete MCMC
• A primer on model-guided exploration of fitness landscapes for biological sequence design
• Autofocused oracles for model-based design
• Deep Extrapolation for Attribute-Enhanced Generation
• Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation
• Data-Driven Offline Optimization For Architecting Hardware Accelerators
• $\Delta$-UQ: Accurate Uncertainty Quantification via Anchor Marginalization