A Visual Analytics System for Improving Attention-based Traffic Forecasting Models

Jin, Seungmin; Lee, Hyunwook; Park, Cheonbok; Chu, Hyeshin; Tae, Yunwon; Choo, Jaegul; Ko, Sungahn

doi:10.1109/tvcg.2022.3209462

Cited by 6 publications

(1 citation statement)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the proliferation of VA systems that aim to increase the explainability of DNNs in multiple tasks, 39,[45][46][47][48][49][50][51] there is a lack of VA systems for prototype-based models. The only VA system that uses prototypes to explain the behavior of neural networks that we are aware of, 24 was not conceived to work with temporal media such as deepfake videos, nor with prototypes that combine spatial and temporal information.…”

Section: Refinement Of Prototype-based Modelsmentioning

confidence: 99%

ProtoExplorer: Interpretable forensic analysis of deepfake videos using prototype exploration and refinement

Bouter,

Pardo,

Geradts

et al. 2024

Information Visualization

View full text Add to dashboard Cite

In high-stakes settings, Machine Learning models that can provide predictions that are interpretable for humans are crucial. This is even more true with the advent of complex deep learning based models with a huge number of tunable parameters. Recently, prototype-based methods have emerged as a promising approach to make deep learning interpretable. We particularly focus on the analysis of deepfake videos in a forensics context. Although prototype-based methods have been introduced for the detection of deepfake videos, their use in real-world scenarios still presents major challenges, in that prototypes tend to be overly similar and interpretability varies between prototypes. This paper proposes a Visual Analytics process model for prototype learning, and, based on this, presents ProtoExplorer, a Visual Analytics system for the exploration and refinement of prototype-based deepfake detection models. ProtoExplorer offers tools for visualizing and temporally filtering prototype-based predictions when working with video data. It disentangles the complexity of working with spatio-temporal prototypes, facilitating their visualization. It further enables the refinement of models by interactively deleting and replacing prototypes with the aim to achieve more interpretable and less biased predictions while preserving detection accuracy. The system was designed with forensic experts and evaluated in a number of rounds based on both open-ended think aloud evaluation and interviews. These sessions have confirmed the strength of our prototype-based exploration of deepfake videos while they provided the feedback needed to continuously improve the system.

show abstract