Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation

02/16/2021
by   Justin Fu, et al.
9

In this work we consider data-driven optimization problems where one must maximize a function given only queries at a fixed set of points. This problem setting emerges in many domains where function evaluation is a complex and expensive process, such as in the design of materials, vehicles, or neural network architectures. Because the available data typically only covers a small manifold of the possible space of inputs, a principal challenge is to be able to construct algorithms that can reason about uncertainty and out-of-distribution values, since a naive optimizer can easily exploit an estimated model to return adversarial inputs. We propose to tackle this problem by leveraging the normalized maximum-likelihood (NML) estimator, which provides a principled approach to handling uncertainty and out-of-distribution inputs. While in the standard formulation NML is intractable, we propose a tractable approximation that allows us to scale our method to high-capacity neural network models. We demonstrate that our method can effectively optimize high-dimensional design problems in a variety of disciplines such as chemistry, biology, and materials engineering.

READ FULL TEXT

page 15

page 17

page 18

page 19

research
12/31/2019

Model Inversion Networks for Model-Based Optimization

In this work, we aim to solve data-driven optimization problems, where t...
research
11/05/2020

Amortized Conditional Normalized Maximum Likelihood

While deep neural networks provide good performance for a range of chall...
research
07/14/2021

Conservative Objective Models for Effective Offline Model-Based Optimization

Computational design problems arise in a number of settings, from synthe...
research
02/17/2022

Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimization

Black-box model-based optimization (MBO) problems, where the goal is to ...
research
05/15/2019

Information criteria for non-normalized models

Many statistical models are given in the form of non-normalized densitie...
research
02/24/2021

A statistical theory of out-of-distribution detection

We introduce a principled approach to detecting out-of-distribution (OOD...
research
08/19/2022

Estimating a potential without the agony of the partition function

Estimating a Gibbs density function given a sample is an important probl...

Please sign up or login with your details

Forgot password? Click here to reset