基於聯邦強化學習的邊緣計算卸載優化方法及通信系統

專利類型

發明

專利國別 (專利申請國家)

中華民國

專利申請案號

110147803

專利證號

I 792784

專利獲證名稱

基於聯邦強化學習的邊緣計算卸載優化方法及通信系統

專利所屬機關 (申請機關)

國立清華大學

獲證日期

2023/02/11

技術說明

【中文】一種基於聯邦強化學習的邊緣計算卸載優化方法及通信系統。此方法是由各用戶設備將網路及任務狀態輸入參與者網路以產生動作權重表，據以選擇動作來執行任務並獲得評價，將相關資料作為經驗輸入回放內存，從中提取多組經驗輸入評價者網路以獲得價值函數，並依序輸入目標參與者及評價者網路進行動作選擇及評價以獲得目標價值函數，用以更新評價者及參與者網路的網路參數，並軟更新目標參與者及評價者網路的網路參數，將學習的平均效益及動作權重表上傳到雲端設備，由雲端設備計算全域權重表後回傳至各用戶設備以更新動作權重表。 【英文】A method and a system for federated reinforcement learning based offloading optimization in edge computing are provided. In the method, each user equipment inputs network and task states into an actor network to generate an actor weighting table, accordingly selects an action for executing the task and obtains an evaluation. The related data is stored as experience in a replay buffer. Some experiences are extracted from the replay buffer, input into a critic network to obtain a value function, and input into a target actor network and a target critic network in order, to obtain a target value function, which are used to update network parameters of the actor and critic networks, and soft update network parameters of the target actor and critic networks. An average utility of learning and the actor weighting table are uploaded to cloud equipment. The cloud equipment accordingly computes a global weighting table and replies the same to user equipment for updating the actor weighting table.

備註

連絡單位 (專責單位/部門名稱)

智財技轉組

連絡電話

03-5715131-62219

網址