AINov 24, 2018

TGE-viz : Transition Graph Embedding for Visualization of Plan Traces and Domains

arXiv:1811.09900v1
Originality Synthesis-oriented
AI Analysis

This addresses the need for better visualization tools in automated planning to reduce cognitive load and improve plan analysis for researchers and practitioners, though it appears incremental as it builds on existing visualization concepts.

The authors tackled the problem of visualizing plan traces in automated planning, which existing methods fail to capture domain structure and dependencies, resulting in high cognitive load for comparison. They introduced TGE-viz, a technique that embeds the transition graph in low-dimensional space, enabling more intuitive visualization and criticism of plans for mixed-initiative planning.

Existing work for plan trace visualization in automated planning uses pipeline-style visualizations, similar to plans in Gantt charts. Such visualization do not capture the domain structure or dependencies between the various fluents and actions. Additionally, plan traces in such visualizations cannot be easily compared with one another without parsing the details of individual actions, which imposes a higher cognitive load. We introduce TGE-viz, a technique to visualize plan traces within an embedding of the entire transition graph of a domain in low dimensional space. TGE-viz allows users to visualize and criticize plans more intuitively for mixed-initiative planning. It also allows users to visually appraise the structure of domains and the dependencies in it.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes