Agentic Discovery with Active Hypothesis Exploration for Visual Recognition

2026-04-14

AI Summary

This work addresses the challenge of efficiently and interpretably discovering neural architectures under resource constraints and in domain-specific settings by proposing a hypothesis-driven neural architecture search framework. The approach models architecture design as a scientific exploration process, leveraging large language models to generate hypotheses and integrating evolutionary branching, multi-agent feedback, and dynamic hypothesis confidence updating to enable cumulative and transferable learning of design principles within a trajectory tree and hypothesis memory bank. Experiments demonstrate that the method substantially improves accuracy on CIFAR-10 from 18.91% to 94.11%, achieves strong generalization across CIFAR-100, Tiny-ImageNet, and MedMNIST, and attains state-of-the-art performance on MedMNIST, thereby validating its paradigm shift from black-box optimization toward interpretable scientific reasoning.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

We introduce HypoExplore, an agentic framework that formulates neural architecture discovery for visual recognition as a hypothesis-driven scientific inquiry. Given a human-specified high-level research direction, HypoExplore ideates, implements, evaluates, and improves neural architectures through evolutionary branching. New hypotheses are created using a large language model by selecting a parent hypothesis to build upon, guided by a dual strategy that balances exploiting validated principles with resolving uncertain ones. Our proposed framework maintains a Trajectory Tree that records the lineage of all proposed architectures, and a Hypothesis Memory Bank that actively tracks confidence scores acquired through experimental evidence. After each experiment, multiple feedback agents analyze the results from different perspectives and consolidate their findings into hypothesis confidence updates. Our framework is tested on discovering lightweight vision architectures on CIFAR-10, with the best achieving 94.11% accuracy evolved from a root node baseline that starts at 18.91%, and generalizes to CIFAR-100 and Tiny-ImageNet. We further demonstrate applicability to a specialized domain by conducting independent architecture discovery runs on MedMNIST, which yield a state-of-the-art performance. We show that hypothesis confidence scores grow increasingly predictive as evidence accumulates, and that the learned principles transfer across independent evolutionary lineages, suggesting that HypoExplore not only discovers stronger architectures, but can help build a genuine understanding of the design space.

Problem

• neural architecture discovery
• visual recognition
• hypothesis-driven exploration
• lightweight architectures
• design space understanding

Innovation

• agentic discovery
• hypothesis-driven exploration
• neural architecture search
• evolutionary branching
• hypothesis memory bank