Abstract
Solution to many real-world problems often involve the use of expert-level knowledge from various specializations. Such inter-disciplinary problems are usually divided into tasks which are then assigned to a set of bots, each specialized in a particular skill. Supervised selection of the right bot each time is cumbersome and not scalable. Hence there is a need for an AI system that identifies the type of task and assigns it to a suitably trained bot. Challenges arise in non-stationary environments when the cost of choosing different bots vary or the bots themselves might evolve in their skills. In this paper, as in Conversational AI, a number of bots are at our disposal, each of which is trained to handle (i.e., answer) a specific type of question in a conversation. We develop a meta-algorithm that learns about capabilities (Skill Discovery) of the available bots in real-time and appropriately selects a relevant bot for the question at hand. We present contextual bandits as a solution in this setting and introduce gradual finetuning of query information to improve Skill Discovery. Using two popular datasets from conversational AI: CoQA and SQuAD, we show promising results of our method on non-stationary environments.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We used ‘BERT-base-uncased’ model from Hugging Face.
References
Code repository for the work presented in this paper. https://github.com/preetigopal/Skill-Discovery
Amazon: How users invoke custom-skills in Alexa (2021). https://developer.amazon.com/en-US/docs/alexa/custom-skills/understanding-how-users-invoke-custom-skills.html, Accessed 07 June 2021
Auer, P., Cesa-Bianchi, N., Fischer, P.: Finite-time analysis of the multiarmed bandit problem. Mach. Learn. 47(2), 235–256 (2002)
Bouneffouf, D., Rish, I., Cecchi, G., Féraud, R.: Context attentive bandits: contextual bandit with restricted context. In: Proceedings of International Joint Conference on Artificial Intelligence, IJCAI 2017, pp. 1468–1475 (2017)
Devlin, J., Chang, M., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding (2018). CoRR abs/1810.04805 http://arxiv.org/abs/1810.04805
Lau, T.T.: When AI becomes a part of our daily lives (2019). https://hbr.org/2019/05/when-ai-becomes-a-part-of-our-daily-lives, Accessed 30 Oct 2021
Li, L., Chu, W., Langford, J., Schapire, R.E.: A contextual-bandit approach to personalized news article recommendation. In: Proceedings of the 19th International Conference on World Wide Web - WWW 2010 (2010)
Lin, B., Bouneffouf, D., Cecchi, G., Rish, I.: Contextual bandit with adaptive feature extraction. arXiv (2020)
Lu, T., Pal, D., Pal, M.: Contextual multi-armed bandits. In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics. Proceedings of Machine Learning Research, vol. 9, pp. 485–492. JMLR Workshop and Conference Proceedings (May 2010)
Rajpurkar, P., Zhang, J., Lopyrev, K., Liang, P.: SQuAD: 100,000+ questions for machine comprehension of text. In: Proceedings of the 2016 Conference on Empirical Methods in NLP, pp. 2383–2392. ACL, Austin (2016)
Rawassizadeh, R., et al.: Manifestation of virtual assistants and robots into daily life: vision and challenges. CCF Trans. Perv. Comput. Interact. 1(3), 163–174 (2019). https://doi.org/10.1007/s42486-019-00014-1
Reddy, S., Chen, D., Manning, C.: CoQA: a conversational question answering challenge. TACL 7, 249–266 (2019)
White, R.W.: Skill discovery in virtual assistants. Commun. ACM 61(11), 106–113 (2018)
Xu, P., Wen, Z., Zhao, H., Gu, Q.: Neural contextual bandits with deep representation and shallow exploration (2020). ArXiv abs/2012.01780 http://arxiv.org/abs/2012.01780
Acknowledgement
This research was partially funded by the Australian Government through the Australian Research Council (ARC). Prof. Venkatesh is the recipient of an ARC Australian Laureate Fellowship (FL170100006).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Gopal, P., Gupta, S., Rana, S., Le, V., Nguyen, T., Venkatesh, S. (2022). Real-Time Skill Discovery in Intelligent Virtual Assistants. In: Gama, J., Li, T., Yu, Y., Chen, E., Zheng, Y., Teng, F. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2022. Lecture Notes in Computer Science(), vol 13280. Springer, Cham. https://doi.org/10.1007/978-3-031-05933-9_25
Download citation
DOI: https://doi.org/10.1007/978-3-031-05933-9_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-05932-2
Online ISBN: 978-3-031-05933-9
eBook Packages: Computer ScienceComputer Science (R0)