基于博弈的機(jī)器人認(rèn)知情感交互模型

黃宏程; 劉寧; 胡敏; 陶洋; 寇蘭

doi:10.11999/JEIT180867

基于博弈的機(jī)器人認(rèn)知情感交互模型

doi: 10.11999/JEIT180867 cstr: 32379.14.JEIT180867

黃宏程^{1, 2},
劉寧¹,
胡敏¹,
陶洋^{1, 2, ,},
寇蘭¹

1.
重慶郵電大學(xué)通信與信息工程學(xué)院重慶 400065
2.
重慶市通信軟件工程技術(shù)研究中心重慶 400065

基金項(xiàng)目: 國(guó)家自然科學(xué)基金(61871062)，重慶郵電大學(xué)科研基金(A2018-07)

詳細(xì)信息

作者簡(jiǎn)介:
黃宏程：男，1979年生，副教授，研究方向?yàn)檎J(rèn)知情感計(jì)算研究、復(fù)雜網(wǎng)絡(luò)與信息傳播理論

劉寧：女，1995年生，碩士生，研究方向?yàn)檎J(rèn)知情感計(jì)算研究

胡敏：女，1971年生，副教授，研究方向?yàn)樾畔⑼ㄐ啪W(wǎng)絡(luò)體系結(jié)構(gòu)、人機(jī)交互理論與技術(shù)應(yīng)用

陶洋：男，1964年生，教授，研究方向?yàn)槿斯ぶ悄?、大?shù)據(jù)與計(jì)算智能

寇蘭：女，1963年生，副教授，研究方向?yàn)镈2D通信、人機(jī)交互理論與技術(shù)應(yīng)用

通訊作者:
陶洋　taoyang@cqupt.edu.cn

中圖分類號(hào): TP242.6
計(jì)量
- 文章訪問(wèn)數(shù): 2986
- HTML全文瀏覽量: 1234
- PDF下載量: 131
- 被引次數(shù): 0
出版歷程
- 收稿日期: 2018-09-02
- 修回日期: 2019-02-26
- 網(wǎng)絡(luò)出版日期: 2019-04-03
- 刊出日期: 2019-10-01

Cognitive Emotion Interaction Model of Robot Based on Game Theory

Hongcheng HUANG^{1, 2},
Ning LIU¹,
Min HU¹,
Yang TAO^{1, 2
, ,},
Lan KOU¹

1.
School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
2.
Chongqing Engineering Research Center of Communication Software, Chongqing 400065, China

Funds: The National Natural Science Foundation of China (61871062), The Scientific Research Foundation of Chongqing University of Posts and Telecommunications (A2018-07)

摘要

摘要: 針對(duì)現(xiàn)有的人機(jī)交互系統(tǒng)普遍存在情感缺失、參與人參與度不高的問(wèn)題，該文依據(jù)PAD情感空間提出一種基于博弈的機(jī)器人認(rèn)知情感交互模型。首先，對(duì)參與人的交互輸入情感進(jìn)行評(píng)估并分析當(dāng)前人機(jī)交互關(guān)系，提取友好度和共鳴度2個(gè)影響因素。其次，模擬人際交往的心理博弈過(guò)程對(duì)參與人和機(jī)器人的情感生成過(guò)程進(jìn)行建模，將嵌入博弈的子博弈完美均衡策略作為機(jī)器人的最優(yōu)情感選擇策略；最后，根據(jù)最優(yōu)情感策略更新機(jī)器人的情感狀態(tài)轉(zhuǎn)移概率，并以6種基本情感的空間坐標(biāo)為標(biāo)簽，得出受到情感刺激后機(jī)器人情感狀態(tài)的空間坐標(biāo)。實(shí)驗(yàn)結(jié)果表明，與其它認(rèn)知交互模型相比，該文模型能夠減少機(jī)器人對(duì)外界情感刺激的依賴并有效引導(dǎo)參與人參與人機(jī)交互，為機(jī)器人的情感認(rèn)知建模提供了新的方法和思路。
- 認(rèn)知情感交互 /
- PAD情感空間 /
- 博弈 /
- 情感策略選擇
Abstract: To solve the problems of the existing in the process of human-computer interaction system, such as lack of emotion and low participation, a cognitive emotion interaction model based on game theory in PAD emotion space is proposed. Firstly, the interactive input emotion of participant is evaluated and some influence factors such as friendship and resonance are extracted to analyze the current human-computer interaction relationship. Secondly, modeling the emotional generation process of participants and robots by simulating the psychological game process in interpersonal communication, and the optimal emotional strategy of the robot is obtained by using the sub-game perfection equilibrium of the embedded game. Finally, the emotional state transition probability of the robot is updated according the optimal emotional strategy. The spatial coordinates of the six basic emotional states are used as labels to obtain the PAD spatial coordinate of the robot emotional state after emotional stimulate, The results of experiment show that compared with the others emotional interaction model, the proposed model can reduce the dependence of robots on external emotional stimuli and effective guide participants to participate in human-computer interaction, which provides some ideas for the emotion cognition model of robot in human-computer interaction.
- Cognitive emotion interaction /
- PAD emotion space /
- Game theory /
- Emotional strategy choice

HTML全文

圖 1 人機(jī)交互過(guò)程

下載: 全尺寸圖片幻燈片

圖 2 機(jī)器人的情感策略選擇過(guò)程

下載: 全尺寸圖片幻燈片

圖 3 不同認(rèn)知模型作用下的機(jī)器人輸出情感與參與人輸入情感的關(guān)聯(lián)度分析

下載: 全尺寸圖片幻燈片

圖 4 參與人與不同認(rèn)知模型作用下的聊天機(jī)器人交互時(shí)的滿意度統(tǒng)計(jì)

下載: 全尺寸圖片幻燈片

表 1 基于博弈的機(jī)器人認(rèn)知情感交互模型構(gòu)建

輸入：$k{{ - 1}}$次會(huì)話后友好度更新值$F(k - 1)$和機(jī)器人的情感狀態(tài)轉(zhuǎn)移概率${{\text{P}}_{\rm{R}}}(k - 1)$, $k$次會(huì)話參與人的交互輸入情感${\text{E}}_{{\rm{HR}}}^k$;
輸出：$k + 1$次會(huì)話時(shí)機(jī)器人的情感值${\text{E}}_{{\rm{RH}}}^{k{{ + 1}}}$;
Repeat:
參與人輸入交互情感${\text{E}}_{{\rm{HR}}}^k$;
根據(jù)式(1)—式(3)將${\text{E}}_{{\rm{HR}}}^k$評(píng)估轉(zhuǎn)化為強(qiáng)度值向量${\text{P}}({\text{E}}_{{\rm{HR}}}^k)$；
根據(jù)式(8)—式(11)計(jì)算針對(duì)$k + 1$次會(huì)話機(jī)器人每種情感策略選擇，預(yù)測(cè)$k + 2$次會(huì)話參與人每種情感策略選擇，$k + 3$次會(huì)話機(jī)器人每種情感策略下參與人和機(jī)器人的效用值；
根據(jù)式(12)，式(13)求解機(jī)器人的情感選擇策略$s$；
通過(guò)最優(yōu)情感策略$s$對(duì)機(jī)器人的情感狀態(tài)轉(zhuǎn)移概率進(jìn)行更新，對(duì)機(jī)器人情感的空間坐標(biāo)進(jìn)行標(biāo)定；
更新人機(jī)交互友好度，并令$k = k + 2$；
Until 參與人停止輸入交互情感；
人機(jī)交互會(huì)話結(jié)束。

下載: 導(dǎo)出CSV

表 2 不同認(rèn)知模型的自動(dòng)評(píng)價(jià)結(jié)果

模型	MRR	MAP
Seq2Seq	0.3836	0.4015
ChatterBot	0.4623	0.4923
MECs	0.5903	0.6091
GCRs	0.6269	0.6435
本文	0.6507	0.6756

下載: 導(dǎo)出CSV

表 3 參與人與不同認(rèn)知模型作用下的機(jī)器人交互的次數(shù)與時(shí)間統(tǒng)計(jì)

機(jī)器人的認(rèn)知模型	平均交互輪數(shù)(輪)	平均交互時(shí)間(s)
Seq2Seq	9	98.32
ChatterBot	6	60.69
MECs	7	88.16
GCRs	10	110.38
本文	12	130.51

下載: 導(dǎo)出CSV

參考文獻(xiàn)(20)

TURKER B B, YEMEZ Y, SEZGIN T M, et al. Audio-facial laughter detection in naturalistic dyadic conversations[J]. IEEE Transactions on Affective Computing, 2017, 8(4): 534–545. doi: 10.1109/TAFFC.2017.2754256

CHEN Min, HERRERA F, and HWANG K. Cognitive computing: Architecture, technologies and intelligent applications[J]. IEEE Access, 2018, 6: 19774–19783. doi: 10.1109/ACCESS.2018.2791469

ZUCCO C, CALABRESE B, and CANNATARO M. Sentiment analysis and affective computing for depression monitoring[C]. The 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Kansas City, USA, 2017: 1988–1995. doi: 10.1109/BIBM.2017.8217966.

BELKAID M, CUPERLIER N, and GAUSSIER P. Autonomous cognitive robots need emotional modulations: Introducing the eMODUL model[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018, 49(1): 206–215. doi: 10.1109/TSMC.2018.2792542

韓晶, 解侖, 劉欣, 等. 基于Gross認(rèn)知重評(píng)的機(jī)器人認(rèn)知情感交互模型[J]. 東南大學(xué)學(xué)報(bào): 自然科學(xué)版, 2015, 45(2): 270–274. doi: 10.3969/j.issn.1001-0505.2015.02.014

HAN Jing, XIE Lun, LIU Xin, et al. Cognitive emotion interaction model of robot based on Gross cognitive reappraisal[J]. Journal of Southeast University:Natural Science Edition, 2015, 45(2): 270–274. doi: 10.3969/j.issn.1001-0505.2015.02.014

LIU Xin, XIE Lun, and WANG Zhiliang. Empathizing with emotional robot based on cognition reappraisal[J]. China Communications, 2017, 14(9): 100–113. doi: 10.1109/CC.2017.8068769

ZHANG Rui, WANG Zhenyu, and MAI Dongcheng. Building emotional conversation systems using multi-task Seq2Seq learning[C]. The Sixth CCF International Conference on Natural Language Processing and Chinese Computing, Dalian, China, 2017: 612–621. doi: 10.1007/978-3-319-73618-1_51.

RODRíGUEZ L F, GUTIERREZ-GARCIA J O, and RAMOS F. Modeling the interaction of emotion and cognition in Autonomous Agents[J]. Biologically Inspired Cognitive Architectures, 2016, 17: 57–70. doi: 10.1016/j.bica.2016.07.008

NANTY A and GELIN R. Fuzzy controlled PAD emotional state of a NAO robot[C]. 2013 Conference on Technologies and Applications of Artificial Intelligence, Taipei, China, 2013: 90–96. doi: 10.1109/TAAI.2013.30.

曹東巖. 基于強(qiáng)化學(xué)習(xí)的開(kāi)放領(lǐng)域聊天機(jī)器人對(duì)話生成算法[D]. [碩士論文], 哈爾濱工業(yè)大學(xué), 2017.

CAO Dongyan. Research on reinforcement learning for open domain chatbot dialogue generation[D]. [Master dissertation], Harbin Institute of Technology, 2017.

ZHOU Hao, HUANG Minlie, ZHANG Tianyang, et al. Emotional chatting machine: Emotional conversation generation with internal and external memory[C]. The Thirty-Second AAAI Conference on Artificial Intelligence, New Orleans, USA, 2018: 730–738.

華生. 欲望心理學(xué): 人際交往中的心理博弈[M]. 北京, 中央編譯出版社, 2016: 1–5.

HUA Sheng. Psychology on Desire: Psychological Game in Interpersonal Communication[M]. Beijing: Central Compilation & Translation Press, 2016: 1–5.

卜湛, 伍之昂, 曹杰, 等. 在線評(píng)論情感計(jì)算與博弈預(yù)測(cè)[J]. 電子學(xué)報(bào), 2015, 43(12): 2530–2535. doi: 10.3969/j.issn.0372-2112.2015.12.028

BU Zhan, WU Zhiang, CAO Jie, et al. Affective computing and game theory based prediction for online reviews[J]. Acta Electronica Sinica, 2015, 43(12): 2530–2535. doi: 10.3969/j.issn.0372-2112.2015.12.028

PARK J W, KIM W H, LEE W H, et al. How to completely use the PAD space for socially interactive robots[C]. 2011 IEEE International Conference on Robotics and Biomimetics, Karon Beach, Thailand, 2011: 3005–3010. doi: 10.1109/ROBIO.2011.6181762.

LI Jiaqi, ZHANG Chunyan, SUN Qinglin, et al. Changing the Intensity of Interaction Based on Individual Behavior in the Iterated Prisoner’s Dilemma Game[J]. IEEE Transactions on Evolutionary Computation, 2017, 21(4): 506–517. doi: 10.1109/TEVC.2016.2628385

MARTINICH L P. Top ten lessons for managers: Deep dive into interpersonal communication[J]. IEEE Engineering Management Review, 2017, 45(2): 27–29. doi: 10.1109/EMR.2017.2701511

SHANG Lifeng, LU Zhengdong, and LI Hang. Neural responding machine for short-text conversation[C]. The 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, Beijing, China, 2015: 1577-1586. doi: 10.3115/v1/p15-1152.

COX G. ChatterBot tutorial[EB/OL]. https://chatterbot.readthedocs.io/en/stable/tutorial.html, 2018.

SUTSKEVER I, VINYALS O, and LE Q V. Sequence to sequence learning with neural networks[C]. The 27th International Conference on Neural Information Processing Systems, Montreal, Canada, 2014: 3104–3112.

WU Yu, WU Wei, XING Chen, et al. Sequential matching network: A new architecture for multi-turn response selection in retrieval-based chatbots[C]. The 55th Annual Meeting of the Association for Computational Linguistics, Vancouver, Canada, 2017: 496–505.

相關(guān)文章

施引文獻(xiàn)

資源附件(0)

訪問(wèn)統(tǒng)計(jì)