日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

軌道予測arXiv:2402.08698

AMEND: 長尾軌道予測のための専門家混合フレームワーク

AMEND: A Mixture of Experts Framework for Long-tailed Trajectory Prediction

シェア:XThreadsFacebookLINEはてブBluesky

歩行者の軌道予測における長尾問題に対処するため、データの各部分に特化した専門家を組み合わせるモジュール型のMixture of Expertsフレームワークを提案し、長尾シナリオでの性能を向上させた。

著者: Ray Coden Mercurius, Ehsan Ahmadi, Soheil Mohamad Alizadeh Shabestary, Amir Rasouli

分類: cs.CV, cs.LG, cs.RO

原文アブストラクト

Accurate prediction of pedestrians' future motions is critical for intelligent driving systems. Developing models for this task requires rich datasets containing diverse sets of samples. However, the existing naturalistic trajectory prediction datasets are generally imbalanced in favor of simpler samples and lack challenging scenarios. Such a long-tail effect causes prediction models to underperform on the tail portion of the data distribution containing safety-critical scenarios. Previous methods tackle the long-tail problem using methods such as contrastive learning and class-conditioned hypernetworks. These approaches, however, are not modular and cannot be applied to many machine learning architectures. In this work, we propose a modular model-agnostic framework for trajectory prediction that leverages a specialized mixture of experts. In our approach, each expert is trained with a specialized skill with respect to a particular part of the data. To produce predictions, we utilise a router network that selects the best expert by generating relative confidence scores. We conduct experimentation on common pedestrian trajectory prediction datasets and show that our method improves performance on long-tail scenarios. We further conduct ablation studies to highlight the contribution of different proposed components.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  3. 3NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  4. 4GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  5. 5Google DeepMind、ロボット向けAIを発表新製品・発表

PR・アフィリエイト広告

MINISFORUM(ミニPC)

MINISFORUM(ミニPC)の公式広告バナーMINISFORUM公式ストアを見る ↗
PR本紙発行元 EmplifAI