Road Network Metric Learning for Estimated Time of Arrival

TL;DR

This paper introduces RNML-ETA, a framework that enhances ETA predictions by addressing data sparsity in road network embeddings through metric learning and a novel triangle loss, improving accuracy especially on cold links.

Contribution

The paper proposes a novel metric learning framework with triangle loss to improve road link embeddings for ETA, tackling data sparsity issues in large-scale networks.

Findings

Outperforms state-of-the-art models on real-world datasets

Improves accuracy on cold links with limited data

Effective in large-scale ride-hailing scenarios

Abstract

Recently, deep learning have achieved promising results in Estimated Time of Arrival (ETA), which is considered as predicting the travel time from the origin to the destination along a given path. One of the key techniques is to use embedding vectors to represent the elements of road network, such as the links (road segments). However, the embedding suffers from the data sparsity problem that many links in the road network are traversed by too few floating cars even in large ride-hailing platforms like Uber and DiDi. Insufficient data makes the embedding vectors in an under-fitting status, which undermines the accuracy of ETA prediction. To address the data sparsity problem, we propose the Road Network Metric Learning framework for ETA (RNML-ETA). It consists of two components: (1) a main regression task to predict the travel time, and (2) an auxiliary metric learning task to improve the quality of link embedding vectors. We further propose the triangle loss, a novel loss function to improve the efficiency of metric learning. We validated the effectiveness of RNML-ETA on large scale real-world datasets, by showing that our method outperforms the state-of-the-art model and the promotion concentrates on the cold links with few data.