research-article

Green Edge Intelligence Scheme for Mobile Keyboard Emoji Prediction

Authors:

Yang YangAuthors Info & Claims

IEEE Transactions on Mobile Computing, Volume 23, Issue 2

Pages 1888 - 1901

https://doi.org/10.1109/TMC.2023.3243955

Published: 01 February 2024 Publication History

Abstract

Emoji prediction has been widely adopted in most mobile keyboards to improve the quality of user experience. Considering the resource constraints of smartphones, it is promising to deploy well-trained prediction models on edge servers, with which smartphones can carry out emoji prediction in an online fashion. However, a key issue in such a scenario lies in how the smartphone should select a subset of models to achieve high-accuracy and real-time emoji prediction with energy efficiency (<italic>a.k.a.</italic> the <italic>model selection</italic> problem). Moreover, part of the system dynamics such as the prediction accuracy and the inference latency of each model are usually unknown <italic>a priori</italic> in practice, further complicating the problem. In this paper, with an effective integration of history-aware online learning and online control, we propose the first green edge intelligence scheme to solve the model selection problem for mobile keyboard emoji prediction. Our theoretical analysis and simulation results verify the effectiveness of our proposed scheme in achieving a sub-linear round-averaged regret bound and energy efficiency with a high prediction accuracy and a low latency.

References

[1]

J. Hou, Y. Tang, X. Huang, Z. Shao, and Y. Yang, “Green edge intelligence scheme for mobile keyboard emoji prediction,” in Proc. IEEE Int. Conf. Commun., 2021, pp. 1–6.

[2]

A. Ramesh et al., “Zero-shot text-to-image generation,” in Proc. Int. Conf. Mach. Learn., 2021, pp. 8821–8831.

[3]

C. Saharia et al., “Photorealistic text-to-image diffusion models with deep language understanding,” 2022,.

[4]

S. Ramaswamy, R. Mathews, K. Rao, and F. Beaufays, “Federated learning for emoji prediction in a mobile keyboard,” 2019,.

[5]

S. Kumar et al., “Voicemoji: A novel on-device pipeline for seamless emoji insertion in dictation,” in Proc. IEEE 18th India Council Int. Conf., 2021, pp. 1–6.

[6]

X. Chen, A. Jindal, N. Ding, Y. C. Hu, M. Gupta, and R. Vannithamby, “Smartphone background activities in the wild: Origin, energy drain, and optimization,” in Proc. 21st Annu. Int. Conf. Mobile Comput. Netw., 2015, pp. 40–52.

[7]

B. Charyyev, E. Arslan, and M. H. Gunes, “Latency comparison of cloud datacenters and edge servers,” in Proc. IEEE Glob. Commun. Conf., 2020, pp. 1–6.

[8]

D. Liu, H. Kong, X. Luo, W. Liu, and R. Subramaniam, “Bringing AI to edge: From deep learning's perspective,” Neurocomputing, vol. 485, pp. 297–320, 2022.

Digital Library

[9]

P. Garcia Lopez et al., “Edge-centric computing: Vision and challenges,” ACM SIGCOMM Comput. Commun. Rev., vol. 45, no. 5, pp. 37–42, 2015.

Digital Library

[10]

iPhone User Guide, “Use predictive text on iPhone,” 2021. [Online]. Available: https://support.apple.com/guide/iphone/use-predictive-text-iphd4ea90231/ios

[11]

P. Mach and Z. Becvar, “Mobile edge computing: A survey on architecture and computation offloading,” IEEE Commun. Surveys Tut., vol. 19, no. 3, pp. 1628–1656, Mar. 2017.

Digital Library

[12]

Y. Feng, L. Guo, X. Fu, and N. Liu, “Efficient mobile energy replenishment scheme based on hybrid mode for wireless rechargeable sensor networks,” IEEE Sensors J., vol. 19, no. 21, pp. 10131–10143, Jul. 2019.

[13]

M. J. Neely, “Stochastic network optimization with application to communication and queueing systems,” Synth. Lectures Commun. Netw., vol. 3, no. 1, pp. 1–211, 2010.

[14]

P. Shivaswamy and T. Joachims, “Multi-armed bandit problems with history,” in Proc. Artif. Intell. Statist., 2012, pp. 1046–1054.

[15]

F. Barbieri, L. E. Anke, J. Camacho-Collados, S. Schockaert, and H. Saggion, “Interpretable emoji prediction via label-wise attention LSTMs,” in Proc. Conf. Empir. Methods Natural Lang. Process., 2018, pp. 4766–4771.

[16]

B. Felbo, A. Mislove, A. Søgaard, I. Rahwan, and S. Lehmann, “Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm,” in Proc. Conf. Empir. Methods Natural Lang. Process., 2017, pp. 1615–1625.

[17]

A. Singh, E. Blanco, and W. Jin, “Incorporating emoji descriptions improves tweet classification,” in Proc. Proc. Conf. North Amer. Assoc. Comput. Linguistics, 2019, pp. 2096–2101.

[18]

P. Zhao, J. Jia, Y. An, J. Liang, L. Xie, and J. Luo, “Analyzing and predicting emoji usages in social media,” in Proc. Int. Conf. World Wide Web, 2018, pp. 327–334.

[19]

D. Li, R. Rzepka, M. Ptaszynski, and K. Araki, “Emoji-aware attention-based Bi-directional GRU network model for chinese sentiment analysis,” in Proc. Int. Joint Conf. Artif. Intell., 2019, pp. 11–18.

[20]

N. Wang, J. Wang, and X. Zhang, “YNU-HPCC at semeval-2018 task 2: Multi-ensemble Bi-GRU model with attention mechanism for multilingual emoji prediction,” in Proc. 12th Int. Workshop Semantic Eval., 2018, pp. 459–465.

[21]

W. Ma, R. Liu, L. Wang, and S. Vosoughi, “Emoji prediction: Extensions and benchmarking,” 2020, arXiv:2007.07389.

[22]

P. Delobelle and B. Berendt, “Time to take emoji seriously: They vastly improve casual conversational models,” in Proc. 31st Benelux Conf. Artif. Intell., 2019, pp. 1–7.

[23]

A. Vaswani et al., “Attention is all you need,” in Proc. Int. Conf. Neural Inf. Process. Syst., 2017, pp. 6000–6010.

[24]

D. Kopev et al., “Tweety at SemEval-2018 task 2: Predicting emojis using hierarchical attention neural networks and support vector machine,” in Proc. 12th Int. Workshop Semantic Eval., 2018, pp. 497–501.

[25]

C. Liebeskind and S. Liebeskind, “Emoji prediction for hebrew political domain,” in Proc. Int. Conf. World Wide Web, 2019, pp. 468–477.

[26]

L. Alexa, A. B. Lorent, D. Gifu, and D. Trandabat, “The dabblers at semeval-2018 task 2: Multilingual emoji prediction,” in Proc. Int. Workshop Semantic Eval., 2018, pp. 405–409.

[27]

A. Nagabandi, G. Kahn, R. S. Fearing, and S. Levine, “Neural network dynamics for model-based deep reinforcement learning with model-free fine-tuning,” in Proc. IEEE Int. Conf. Robot. Automat., 2018, pp. 7559–7566.

[28]

J. Beaulieu and D. Asamoah Owusu, “Umduluth-cs8761 at SemEval-2018 task 2: Emojis: Too many choices?,” in Proc. Int. Workshop Semantic Eval., 2018, pp. 400–404.

[29]

J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” 2018,.

[30]

S. Yu, N. Kulkarni, H. Lee, and J. Kim, “On-device neural language model based word prediction,” in Proc. Int. Conf. Comput. Linguistics, 2018, pp. 128–131.

[31]

A. Hard et al., “Federated learning for mobile keyboard prediction,” 2018,.

[32]

Y. Xu and S. Mao, “A survey of mobile cloud computing for rich media applications,” IEEE Wirel. Commun., vol. 20, no. 3, pp. 46–53, Jun. 2013.

[33]

D. J. Foster, A. Krishnamurthy, and H. Luo, “Model selection for contextual bandits,” in Proc. Int. Conf. Neural Inf. Process. Syst., 2019, pp. 14741–14752.

[34]

A. Pacchiano et al., “Model selection in contextual stochastic bandit problems,” in Proc. Int. Conf. Neural Inf. Process. Syst., 2020, pp. 10328–10337.

[35]

C. Z. Felício, K. V. Paixão, C. A. Barcelos, and P. Preux, “A multi-armed bandit model selection for cold-start user recommendation,” in Proc. User Model., Adapt. Personalization, 2017, pp. 32–40.

[36]

J. Xie, M. Tashman, J. Hoffman, L. Winikor, and R. Gerami, “Online and scalable model selection with multi-armed bandits,” 2021, arXiv:2101.10385.

[37]

N. Shukla, A. Kolbeinsson, L. Marla, and K. Yellepeddi, “Adaptive model selection framework: An application to airline pricing,” 2019,.

[38]

V. Pejovic and M. Musolesi, “Anticipatory mobile computing: A survey of the state of the art and research challenges,” Comput. Surv., vol. 47, no. 3, pp. 1–29, 2015.

Digital Library

[39]

S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Comput., vol. 9, no. 8, pp. 1735–1780, 1997.

Digital Library

[40]

A. Mylonas, V. Meletiadis, B. Tsoumas, L. Mitrou, and D. Gritzalis, “Smartphone forensics: A proactive investigation scheme for evidence acquisition,” in Proc. IFIP Int. Inf. Secur. Conf., 2012, pp. 249–260.

[41]

W. Chen, Y. Wang, and Y. Yuan, “Combinatorial multi-armed bandit: General framework, results and applications,” in Proc. Int. Conf. Mach. Learn., 2013, pp. 151–159.

[42]

A. Slivkins et al., “Introduction to multi-armed bandits,” Found. Trends Mach. Learn., vol. 12, no. 1/2, pp. 1–286, 2019.

Digital Library

[43]

F. Li, J. Liu, and B. Ji, “Combinatorial sleeping bandits with fairness constraints,” in Proc. IEEE Conf. Comput. Commun., 2019, pp. 1702–1710.

[44]

W. Hoeffding, “Probability inequalities for sums of bounded random variables,” J. Amer. Statist. Assoc., vol. 58, no. 301, pp. 13–30, 1963.

[45]

J. L. W. V. Jensen, “Sur les fonctions convexes et les inégalités entre les valeurs moyennes,” Acta Mathematica, vol. 30, no. 1, pp. 175–193, 1906.

[46]

F. Barbieri et al., “Semeval 2018 task 2: Multilingual emoji prediction,” in Proc. Int. Workshop Semantic Eval., 2018, pp. 24–33.

[47]

J. Platt et al., “Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods,” Adv. Large Margin Classifiers, vol. 10, no. 3, pp. 61–74, 1999.

[48]

W.-Y. Loh, “Classification and regression trees,” Wiley Interdiscipl. Rev.: Data Mining Knowl. Discov., vol. 1, no. 1, pp. 14–23, 2011.

[49]

A. Sherstinsky, “Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network,” Physica D: Nonlinear Phenomena, vol. 404, 2020, Art. no.

[50]

Y. Liu et al., “RoBERTa: A robustly optimized BERT pretraining approach,” 2019,.

[51]

Y. Gong, L. Liu, M. Yang, and L. Bourdev, “Compressing deep convolutional networks using vector quantization,” 2014,.

Cited By

Huang XTang YLi JZhang NShen X(2024)Toward Effective Retrieval Augmented Generative Services in 6G NetworksIEEE Network: The Magazine of Global Internetworking10.1109/MNET.2024.343667038:6(459-467)Online publication date: 1-Nov-2024
https://dl.acm.org/doi/10.1109/MNET.2024.3436670

Index Terms

Green Edge Intelligence Scheme for Mobile Keyboard Emoji Prediction
1. Computing methodologies
  1. Machine learning
2. Human-centered computing

Index terms have been assigned to the content through auto-classification.

Recommendations

The Emoji Spatial Stroop Task: Exploring the impact of vertical positioning of emoji on emotional processing
Abstract
Despite emoji often being assumed to be a form of emotional communication, the emotional affordances of these are not yet fully established. The current study employed the Emoji Spatial Stroop Task to explore whether spatial iconicity ...
Highlights
- Emoji Spatial Stroop Task measures spatial iconicity on semantic-related judgments.
A novel green learning artificial intelligence model for regional electrical load prediction
Abstract
Load prediction is crucial to unit commitment and scheduling in electricity systems; however, load prediction models consume considerable electricity. Most conventional prediction methods are based on deep learning (DL) models, such as long short-...
Tourism Destination Preference Prediction Based on Edge Computing
In view of the fact that the important characteristics of tourism destination selection preference are not considered in the current prediction methods of tourism destination selection preference, resulting in low prediction accuracy and comprehensive ...

Comments

Information & Contributors

Information

Published In

cover image IEEE Transactions on Mobile Computing

IEEE Transactions on Mobile Computing Volume 23, Issue 2

Feb. 2024

1002 pages

Issue’s Table of Contents

1536-1233 © 2023 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.

Publisher

IEEE Educational Activities Department

United States

Publication History

Published: 01 February 2024

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

1
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 01 Feb 2025

Other Metrics

View Author Metrics

Citations

Cited By

Huang XTang YLi JZhang NShen X(2024)Toward Effective Retrieval Augmented Generative Services in 6G NetworksIEEE Network: The Magazine of Global Internetworking10.1109/MNET.2024.343667038:6(459-467)Online publication date: 1-Nov-2024
https://dl.acm.org/doi/10.1109/MNET.2024.3436670

View Options

View options

Figures

Tables

Media

View Issue’s Table of Contents