# Novelty-Driven Target-Space Discovery in Automated Electron and Scanning Probe Microscopy

arXiv: https://arxiv.org/abs/2603.16715 · HTML: https://mlsanity.com/paper/2603.16715/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 65 references that are in the map.
- First posted on arXiv 17 Mar 2026.

## Builds on (most cited first, 20 of 20)

- [GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration](https://mlsanity.com/paper/1809.11165/) (2018)
- [Deep Kernel Learning](https://mlsanity.com/paper/1511.02222/) (2015)
- [A Survey of Human-in-the-loop for Machine Learning](https://mlsanity.com/paper/2108.00941/) (2021)
- [BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization](https://mlsanity.com/paper/1910.06403/) (2019)
- [Bayesian Active Learning for Scanning Probe Microscopy: from Gaussian Processes to Hypothesis Learning](https://mlsanity.com/paper/2205.15458/) (2022)
- [Machine Learning Pipeline for Segmentation and Defect Identification from High Resolution Transmission Electron Microscopy Data](https://mlsanity.com/paper/2001.05022/) (2020)
- [Human-in-the-loop: The future of Machine Learning in Automated Electron Microscopy](https://mlsanity.com/paper/2310.05018/) (2023)
- [A Deep Learning Based Automatic Defect Analysis Framework for In-situ TEM Ion Irradiations](https://mlsanity.com/paper/2108.08882/) (2021)
- [Deep Kernel Methods Learn Better: From Cards to Process Optimization](https://mlsanity.com/paper/2303.14554/) (2023)
- [Integration of Scanning Probe Microscope with High-Performance Computing: fixed-policy and reward-driven workflows implementation](https://mlsanity.com/paper/2405.12300/) (2024)
- [SANE: Strategic Autonomous Non-Smooth Exploration for Multiple Optima Discovery in Multi-modal and Non-differentiable Black-box Functions](https://mlsanity.com/paper/2409.12295/) (2024)
- [Active learning in open experimental environments: selecting the right information channel(s) based on predictability in deep kernel learning](https://mlsanity.com/paper/2203.10181/) (2022)
- [Unraveling the Impact of Initial Choices and In-Loop Interventions on Learning Dynamics in Autonomous Scanning Probe Microscopy](https://mlsanity.com/paper/2402.00071/) (2024)
- [Building Workflows for Interactive Human in the Loop Automated Experiment (hAE) in STEM-EELS](https://mlsanity.com/paper/2404.07381/) (2024)
- [Co-orchestration of Multiple Instruments to Uncover Structure-Property Relationships in Combinatorial Libraries](https://mlsanity.com/paper/2402.02198/) (2024)
- [Beyond Optimization: Exploring Novelty Discovery in Autonomous Experiments](https://mlsanity.com/paper/2508.20254/) (2025)
- [Integrating Predictive and Generative Capabilities by Latent Space Design via the DKL-VAE Model](https://mlsanity.com/paper/2503.02978/) (2025)
- [BEACON: A Bayesian Optimization Inspired Strategy for Efficient Novelty Search](https://mlsanity.com/paper/2406.03616/) (2024)
- [Domain Switching on the Pareto Front: Multi-Objective Deep Kernel Learning in Automated Piezoresponse Force Microscopy](https://mlsanity.com/paper/2506.08073/) (2025)
- [Towards Self-Optimizing Electron Microscope: Robust Tuning of Aberration Coefficients via Physics-Aware Multi-Objective Bayesian Optimization](https://mlsanity.com/paper/2601.18972/) (2026)

## Abstract (arXiv)

Modern automated microscopy faces a fundamental discovery challenge: in many systems, the most important scientific information does not reside in the immediately visible image features, but in the target space of sequentially acquired spectra or functional responses, making it essential to develop strategies that can actively search for new behaviors rather than simply optimize known objectives. Here, we developed a deep-kernel-learning BEACON framework that is explicitly designed to guide discovery in the target space by learning structure-property relationships during the experiment and using that evolving model to seek diverse response regimes. We first established the method through demonstration workflows built on pre-acquired ground-truth datasets, which enabled direct benchmarking against classical acquisition strategies and allowed us to define a set of monitoring functions for comparing exploration quality, target-space coverage, and surrogate-model behavior in a transparent and reproducible manner. This benchmarking framework provides a practical basis for evaluating discovery-driven algorithms, not just optimization performance. We then operationalized and deployed the workflow on STEM, showing that the approach can transition from offline validation to real experimental implementation. To support adoption and extension by the broader community, the associated notebooks are available, allowing users to reproduce the workflows, test the benchmarks, and adapt the method to their own instruments and datasets.