Action Recognition and State Change Prediction in a Recipe Understanding Task Using a Lightweight Neural Network Model

被引:0
|
作者
Wan, Qing [1 ]
Choe, Yoonsuck [1 ]
机构
[1] Texas A&M Univ, Dept Comp Sci & Engn, College Stn, TX 77843 USA
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Consider a natural language sentence describing a specific step in a food recipe. In such instructions, recognizing actions (such as press, bake, etc.) and the resulting changes in the state of the ingredients (shape molded, custard cooked, temperature hot, etc.) is a challenging task. One way to cope with this challenge is to explicitly model a simulator module that applies actions to entities and predicts the resulting outcome (Bosselut et al. 2018). However, such a model can be unnecessarily complex. In this paper, we propose a simplified neural network model that separates action recognition and state change prediction, while coupling the two through a novel loss function. This allows learning to indirectly influence each other. Our model, although simpler, achieves higher state change prediction performance (67% average accuracy for ours vs. 55% in (Bosselut et al. 2018)) and takes fewer samples to train (10K ours vs. 65K+ by (Bosselut et al. 2018)).
引用
收藏
页码:13945 / 13946
页数:2
相关论文
共 50 条
  • [41] KNOWLEDGE-GUIDED RECURRENT NEURAL NETWORK LEARNING FOR TASK-ORIENTED ACTION PREDICTION
    Lin, Liang
    Huang, Lili
    Chen, Tianshui
    Gan, Yukang
    Cheng, Hui
    2017 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO (ICME), 2017, : 625 - 630
  • [42] Using generalized regression neural network (GRNN) for mechanical strength prediction of lightweight mortar
    Razavi, S. V.
    Jumaat, M. Z.
    Ahmed, E. S. H.
    Mohammadi, P.
    COMPUTERS AND CONCRETE, 2012, 10 (04): : 379 - 390
  • [43] Analysis of the sports action recognition model based on the LSTM recurrent neural network
    Chen, Ping
    Peng, Jiangui
    NONLINEAR ENGINEERING - MODELING AND APPLICATION, 2025, 14 (01):
  • [44] A Lightweight Siamese Neural Network for Building Change Detection Using Remote Sensing Images
    Yang, Haiping
    Chen, Yuanyuan
    Wu, Wei
    Pu, Shiliang
    Wu, Xiaoyang
    Wan, Qiming
    Dong, Wen
    REMOTE SENSING, 2023, 15 (04)
  • [45] Computational Model Based on Neural Network of Visual Cortex for Human Action Recognition
    Liu, Haihua
    Shu, Na
    Tang, Qiling
    Zhang, Wensheng
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (05) : 1427 - 1440
  • [46] A Novel Lightweight Deep Convolutional Neural Network Model for Human Emotions Recognition in Diverse Environments
    Kalsum, Tehmina
    Mehmood, Zahid
    JOURNAL OF SENSORS, 2023, 2023
  • [47] Pulsar Candidate Recognition Using Deep Neural Network Model
    Yin, Qian
    Wang, Yan
    Zheng, Xin
    Zhang, Jikai
    ELECTRONICS, 2022, 11 (14)
  • [48] Visual Shape Recognition Neural Network Using BESOM Model
    Hasegawa, Hiroaki
    Hagiwara, Masafumi
    ARTIFICIAL NEURAL NETWORKS (ICANN 2010), PT III, 2010, 6354 : 102 - 105
  • [49] An Improved Promoter Recognition Model Using Convolutional Neural Network
    Qian, Ying
    Zhang, Yu
    Guo, Binyu
    Ye, Shasha
    Wu, Yuzhu
    Zhang, Jiongmin
    2018 IEEE 42ND ANNUAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE (COMPSAC), VOL 1, 2018, : 471 - 476
  • [50] Spectral recognition using a modified Eckhorn neural network model
    Rughooputh, H.C.S.
    Rughooputh, S.D.D.V.
    Image and Vision Computing, 2001, 19 (1-2) : 1101 - 1103