Neural Network Exploration Using Optimal Experiment Design

I consider the question "How should one act when the only goal is to learn as much as possible?". Building on the theoretical results of Fedorov (1972, Theory of Optimal Experiments, Academic Press) and MacKay (1992, Neural Computation, 4, 590-604), I apply techniques from optimal experiment design (OED) to guide the query/action selection of a neural network learner. I demonstrate that these techniques allow the learner to minimize its generalization error by exploring its domain efficiently and completely. I conclude that, while not a panacea, OED-based query/action selection has much to offer, especially in domains where its high computational costs can be tolerated. Copyright 1996 Elsevier Science Ltd

• Publication year

  1993

• Venue

  Neural Information Processing Systems

• Publication date

  1993-11-29

• Fields of study

  Medicine, Computer Science

• Identifiers
  DOI 10.1016/0893-6080(95)00137-9PMID 12662584
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Active Learning with Statistical Models

  1996cited by this paper
• A designed experiment on the use of neural-network models in short-term hourly load forecasting /

  1995cited by this paper
• Learning to Act Using Real-Time Dynamic Programming

  1995cited by this paper
• REINFORCEMENT DRIVEN INFORMATION ACQUISITION IN NON-DETERMINISTIC ENVIRONMENTS

  1995cited by this paper
• Computing second derivatives in feed-forward networks: a review

  1994cited by this paper
• Query construction, entropy, and generalization in neural-network models.

  1994influential reference
• Robot juggling: implementation of memory-based learning

  1994cited by this paper
• Optimum experimental designs

  1994influential reference
• Bayesian Query Construction for Neural Network Models

  1994cited by this paper
• Fast Exact Multiplication by the Hessian

  1994cited by this paper
• Supervised learning from incomplete data via an EM approach

  1993cited by this paper
• Selecting concise training sets from clean data

  1993cited by this paper
• Forward Models: Supervised Learning with a Distal Teacher

  1992cited by this paper
• Information, Prediction, and Query by Committee

  1992cited by this paper
• How Tight Are the Vapnik-Chervonenkis Bounds?

  1992cited by this paper
• Neural Networks and the Bias/Variance Dilemma

  1992cited by this paper
• Decision Theoretic Generalizations of the PAC Model for Neural Net and Other Learning Applications

  1992cited by this paper
• Information-Based Objective Functions for Active Data Selection

  1992influential reference
• The role of exploration in learning control

  1992cited by this paper
• Introduction to robotics

  1991cited by this paper
• Active Exploration in Dynamic Environments

  1991cited by this paper
• Adaptive Mixtures of Local Experts

  1991cited by this paper
• Soft competitive adaptation: neural network learning algorithms based on fitting statistical mixtures

  1991cited by this paper
• A general regression neural network

  1991cited by this paper
• Constructing Hidden Units Using Examples and Queries

  1990influential reference
• On the sample complexity of pac-learning using random and chosen examples

  1990cited by this paper
• Connectionist Robot Motion Planning: A Neurally-Inspired Approach to Visually-Guided Reaching

  1990cited by this paper
• Query learning based on boundary search and gradient computation of trained multilayer perceptrons

  1990cited by this paper
• Mixture Models: Inference and Applications to Clustering

  1989cited by this paper
• Learnability and the Vapnik-Chervonenkis dimension

  1989cited by this paper
• Training Connectionist Networks with Queries and Selective Sampling

  1989cited by this paper
• Recent advances in nonlinear experiment design

  1989cited by this paper
• On Finding Exciting Trajectories for Identification Experiments Involving Systems with Nonlinear Dynamics

  1989cited by this paper
• What Size Net Gives Valid Generalization?

  1989cited by this paper
• Regression by local fitting: Methods, properties, and computational algorithms

  1988cited by this paper
• Neural model of adaptive hand-eye coordination for single postures.

  1988cited by this paper
• Elements of Statistical Computing: Numerical Computation

  1988cited by this paper
• Empirical Model‐Building and Response Surfaces

  1988cited by this paper
• Elements of statistical computing

  1986cited by this paper
• Introduction to robotics

  1985cited by this paper
• A theory of the learnable

  1984cited by this paper
• A Note on the Number of Queries Needed to Identify Regular Languages

  1981cited by this paper
• Theory Of Optimal Experiments

  1972cited by this paper

• Generative Bayesian Computation as a Scalable Alternative to Gaussian Process Surrogates

  2026cites this paper
• BayesFusion-SDF: Probabilistic Signed Distance Fusion with View Planning on CPU

  2026cites this paper
• Hybrid active learning with wavelet scattering transform and attention-augmented networks for underwater acoustic signal classification under data scarcity

  2026cites this paper
• Active Learning for Manifold Gaussian Process Regression

  2025cites this paper
• Enhancing active learning through latent space exploration: A k-nearest neighbors approach

  2025cites this paper
• Distributionally Robust Active Learning for Gaussian Process Regression

  2025cites this paper
• Deep Intrinsic Coregionalization Multi-Output Gaussian Process Surrogate with Active Learning

  2025cites this paper
• Apax: A Flexible and Performant Framework for the Development of Machine-Learned Interatomic Potentials

  2025cites this paper
• Sequential design for the efficient estimation of offshore structure failure probability

  2025cites this paper
• Gaussian Process Assisted Meta-learning for Image Classification and Object Detection Models

  2025cites this paper
• Cluster-aware threshold optimization with active learning for multi-fidelity surrogates

  2025cites this paper
• Automated Capability Evaluation of Foundation Models

  2025cites this paper
• Fully Bayesian Sequential Design for Mean Response Surface Prediction of Heteroscedastic Stochastic Simulations

  2025cites this paper
• Uncertainty in the era of machine learning for atomistic modeling

  2025cites this paper
• SYSAI for System Health Management - a Statistical Framework for the Analysis of Diagnosis Systems

  2024cites this paper
• Optimal Experimental Design for Universal Differential Equations

  2024cites this paper
• Optimizing machine learning yield functions using query-by-committee for support vector classification with a dynamic stopping criterion

  2024cites this paper
• Query Learning on the Semisupervised Margin

  2024cites this paper
• A meta-active learning approach exploiting instance importance

  2024cites this paper
• ALPBench: A Benchmark for Active Learning Pipelines on Tabular Data

  2024cites this paper
• Data‐Driven Parameterization Refinement for the Structural Optimization of Cruise Ship Hulls

  2024cites this paper
• Active learning for adaptive surrogate model improvement in high-dimensional problems

  2024cites this paper
• A Unified Gaussian Process for Branching and Nested Hyperparameter Optimization

  2024cites this paper
• A prediction rigidity formalism for low-cost uncertainties in trained neural networks

  2024cites this paper
• Non‐uniform active learning for Gaussian process models with applications to trajectory informed aerodynamic databases

  2024cites this paper
• Active Learning for a Recursive Non-Additive Emulator for Multi-Fidelity Computer Experiments

  2023influential citation
• Active Learning and Bayesian Optimization: A Unified Perspective to Learn with a Goal

  2023cites this paper
• Batch-Mode Active Learning of Gaussian Process Regression With Maximum Model Change

  2023cites this paper
• A Meta-Active Learning Approach Exploiting Instance Importance based on Learning Gradient Variation

  2023cites this paper
• Robust optimal designs using a model misspecification term

  2023cites this paper
• Sparse Inverse Cholesky Factorization of Dense Kernel Matrices by Greedy Conditional Selection

  2023influential citation
• Multi-Robot Adaptive Sampling based on Mixture of Experts Approach to Modeling Non-Stationary Spatial Fields

  2023cites this paper
• Sparse Cholesky factorization by greedy conditional selection

  2023cites this paper
• FLEX: an Adaptive Exploration Algorithm for Nonlinear Systems

  2023cites this paper
• Contour location for reliability in airfoil simulation experiments using deep Gaussian processes

  2023cites this paper
• Statistical Analysis and Runtime Monitoring for an AI-based Autonomous Centerline Tracking System

  2023cites this paper
• Predicting the response of complex systems for coastal management

  2023cites this paper
• Prediction-Oriented Bayesian Active Learning

  2023cites this paper
• Gaussian process autoregression models for the evolution of polycrystalline microstructures subjected to arbitrary stretching tensors

  2023cites this paper
• Self-Correcting Bayesian Optimization through Bayesian Active Learning

  2023cites this paper
• Augmenting a Simulation Campaign for Hybrid Computer Model and Field Data Experiments

  2023cites this paper
• Training a Neural-Network-Based Surrogate Model for Aerodynamic Optimisation Using a Gaussian Process

  2022influential citation
• Active Learning for Contextual Search with Binary Feedback

  2022cites this paper
• Traditional kriging versus modern Gaussian processes for large‐scale mining data

  2022cites this paper
• Reactive Exploration to Cope with Non-Stationarity in Lifelong Reinforcement Learning

  2022cites this paper
• Gaussian Processes for the Analysis of Electrochemical Impedance Spectroscopy Data: Prediction, Filtering, and Active Learning

  2022cites this paper
• Machine learning and geospatial methods for large-scale mining data

  2022cites this paper
• A design-space minimum sampling strategy based on proper-orthogonal-decomposition projection

  2022cites this paper
• Toward Runtime Assurance of Complex Systems with AI Components

  2022cites this paper
• Environmental sensor placement with convolutional Gaussian neural processes

  2022cites this paper
• Deep Active Learning with Noise Stability

  2022cites this paper
• The Effect of Task Ordering in Continual Learning

  2022cites this paper
• Sequential adaptive design for emulating costly computer codes

  2022cites this paper
• Introspective Learning : A Two-Stage Approach for Inference in Neural Networks

  2022influential citation
• A Framework and Benchmark for Deep Batch Active Learning for Regression

  2022cites this paper
• Learning to Active Learn by Gradient Variation based on Instance Importance

  2022cites this paper
• Active Sampling: A Machine-Learning-Assisted Framework for Finite Population Inference with Optimal Subsamples

  2022cites this paper
• Optimizing Data Collection for Machine Learning

  2022cites this paper
• Output-Weighted Optimal Sampling for Bayesian Experimental Design and Uncertainty Quantification

  2021cites this paper
• Multiple-criteria Based Active Learning with Fixed-size Determinantal Point Processes

  2021cites this paper
• Active learning for imbalanced data under cold start

  2021cites this paper
• Entropy-based adaptive design for contour finding and estimating reliability

  2021cites this paper
• Active learning for online training in imbalanced data streams under cold start

  2021cites this paper
• Active Output Selection for an HEV Boost Maneuver

  2021cites this paper
• Statement of originality

  2021cites this paper
• Active Learning for Contextual Search with Binary Feedbacks

  2021cites this paper
• Adaptive Local Kernels Formulation of Mutual Information with Application to Active Post-Seismic Building Damage Inference

  2021cites this paper
• A Dataset Perspective on Offline Reinforcement Learning

  2021cites this paper
• Exploration of transferable and uniformly accurate neural network interatomic potentials using optimal experimental design

  2021cites this paper
• Statistical learning framework for safety and failure analysis of a DNN-based autonomous aircraft system

  2021cites this paper
• Active Learning with Maximum Margin Sparse Gaussian Processes

  2021cites this paper
• Global Variance as a Utility Function in Bayesian Optimization

  2021cites this paper
• Improved active output selection strategy for noisy environments

  2021influential citation
• Stream-based active learning for sliding windows under the influence of verification latency

  2021cites this paper
• Sensitivity Prewarping for Local Surrogate Modeling

  2021cites this paper
• Machine learning approaches for feature engineering of the crystal structure: Application to the prediction of the formation energy of cubic compounds

  2021cites this paper
• Recursive partitioning and Gaussian Process Regression for the detection and localization of damages in pultruded Glass Fiber Reinforced Polymer material

  2021cites this paper
• Feature Engineering for Microstructure–Property Mapping in Organic Photovoltaics

  2021cites this paper
• Large-scale local surrogate modeling of stochastic simulation experiments

  2021cites this paper
• Leveraging Local Variation in Data: Sampling and Weighting Schemes for Supervised Deep Learning

  2021cites this paper
• Generalized Chernoff Sampling for Active Testing, Active Regression and Structured Bandit Algorithms

  2021cites this paper
• A Comparative Survey: Benchmarking for Pool-based Active Learning

  2021cites this paper
• Output-Weighted Importance Sampling for Bayesian Experimental Design and Uncertainty Quantification

  2020cites this paper
• Multi-Fingered Active Grasp Learning

  2020cites this paper
• Informative Path Planning for Anomaly Detection in Environment Exploration and Monitoring

  2020cites this paper
• Locally induced Gaussian processes for large-scale simulation experiments

  2020influential citation
• Skill Learning by Autonomous Robotic Playing Using Active Learning and Exploratory Behavior Composition

  2020cites this paper
• Precision aggregated local models

  2020cites this paper
• Active machine learning for spatio-temporal predictions using feature embedding

  2020cites this paper
• Active Learning for Deep Gaussian Process Surrogates

  2020influential citation
• A Framework for the Analysis of Deep Neural Networks in Aerospace applications using Bayesian Statistics

  2020cites this paper
• Local Linear Neuro-Fuzzy Models: Advanced Aspects

  2020cites this paper
• Fast Approximation of the Gauss-Newton Hessian Matrix for the Multilayer Perceptron

  2020cites this paper
• Model Elicitation through Direct Questioning

  2020cites this paper
• Strategies for sequential design of experiments and augmentation

  2020cites this paper
• A Framework for Software Health Management using Bayesian Statistics: Position Paper

  2020cites this paper
• Accelerated knowledge discovery from omics data by optimal experimental design

  2020cites this paper
• Chernoff Sampling for Active Testing and Extension to Active Regression

  2020cites this paper
• A Framework for the Analysis of Deep Neural Networks in Autonomous Aerospace Applications using Bayesian Statistics

  2020cites this paper
• Active Output Selection Strategies for Multiple Learning Regression Models

  2020influential citation