Filters








2,981 Hits in 5.5 sec

A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions [article]

Eric Schulz, Maarten Speekenbrink, Andreas Krause
2016 bioRxiv   pre-print
regression problems, a demonstration of kernel-encoded prior assumptions and compositions, a pure exploration scenario within an optimal design framework, and a bandit-like exploration-exploitation scenario  ...  This tutorial introduces the reader to Gaussian process regression as an expressive tool to model, actively explore and exploit unknown functions.  ...  movies As an example of applying GP-UCB to exploration-exploitation problems, we will use it in an movie recommendation scenario, where the task is to recommend the best movies possible to a user with  ... 
doi:10.1101/095190 fatcat:6ikskyod5fgddcvkzucwrbrwxm

The Externalities of Exploration and How Data Diversity Helps Exploitation [article]

Manish Raghavan, Aleksandrs Slivkins, Jennifer Wortman Vaughan, Zhiwei Steven Wu
2018 arXiv   pre-print
Online learning algorithms, widely used to power search and content optimization on the web, must balance exploration and exploitation, potentially sacrificing the experience of current users for information  ...  bandits model.  ...  Acknowledgments We thank Solon Barocas, Dylan Foster, Jon Kleinberg, Aaron Roth, and Hanna Wallach for helpful discussions about these topics.  ... 
arXiv:1806.00543v2 fatcat:cnfrvcwp7nff3ke4y423ygtwfy

Enhancing Collaborative Filtering Music Recommendation By Balancing Exploration And Exploitation

Zhe Xing, Xinxi Wang, Ye Wang
2014 Zenodo  
This project is funded by the National Research Foundation (NRF) and managed through the multi-agency Interactive & Digital Media Programme Office (IDMPO) hosted by the Media Development Authority of Singapore  ...  With the appropriate amount of exploration, the recommender system could gain more knowledge about the user's true preferences before exploiting them.  ...  It is thus fitting to treat music recommendation as a well-studied reinforcement learning task called n-armed bandit. The n-armed bandit problem assumes a slot machine with n levers.  ... 
doi:10.5281/zenodo.1416775 fatcat:bgzyxysmyrg4zo2k7htdaijqxu

Learning to trade off between exploration and exploitation in multiclass bandit prediction

Hamed Valizadegan, Rong Jin, Shijun Wang
2011 Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '11  
Banditron [8], a multiclass online learning algorithm for bandit setting, maximizes the run-time gain by balancing between exploration and exploitation with a fixed tradeoff parameter.  ...  The exploration vs. exploitation tradeoff strategy is a well-known technique for online learning with incomplete feedback (i.e., bandit setup).  ...  The exploitation vs. exploration tradeoff strategy was originally designed for the multi-armed bandit problem [15] where a slot machine has multiple arms and the player needs to choose the arm with minimum  ... 
doi:10.1145/2020408.2020445 dblp:conf/kdd/ValizadeganJW11 fatcat:yyc43xwunzclfayhktnfr5tu6y

A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions

Eric Schulz, Maarten Speekenbrink, Andreas Krause
2018 Journal of Mathematical Psychology  
regression problems, a demonstration of kernel-encoded prior assumptions and compositions, a pure exploration scenario within an optimal design framework, and a bandit-like exploration-exploitation scenario  ...  This tutorial introduces the reader to Gaussian process regression as an expressive tool to model, actively explore and exploit unknown functions.  ...  Within an exploration-exploitation context, Borji and Itti (2013) and Wu, Schulz, Speekenbrink, Nelson, and Meder (2017) showed that Gaussian process-based optimization can explain how participants  ... 
doi:10.1016/j.jmp.2018.03.001 fatcat:uys4bzhqurhqfll56kzf5oii2u

Exploration, Exploitation, and Engagement in Multi-Armed Bandits with Abandonment [article]

Zixian Yang, Xin Liu, Lei Ying
2022 arXiv   pre-print
Multi-armed bandit (MAB) is a classic model for understanding the exploration-exploitation trade-off.  ...  We propose two algorithms, ULCB and KL-ULCB, both of which do more exploration (being optimistic) when the user likes the previous recommended item and less exploration (being pessimistic) when the user  ...  The system interacts with the users to learn their preferences and recommends personalized contents (learning subjects, videos, songs, products etc.) to each user.  ... 
arXiv:2205.13566v1 fatcat:ngve5ruj7zgpplaxmhjqqlb52a

Explainable Recommendation: A Survey and New Perspectives [article]

Yongfeng Zhang, Xu Chen
2020 arXiv   pre-print
Explainable recommendation helps to improve the transparency, persuasiveness, effectiveness, trustworthiness, and satisfaction of recommendation systems.  ...  We end the survey by discussing potential future directions to promote the explainable recommendation research area and beyond.  ...  Any opinions, findings and conclusions expressed in this material are those of the authors and do not necessarily reflect those of the sponsors.  ... 
arXiv:1804.11192v10 fatcat:scsd3htz65brbiae35zd3nixe4

Information Structures for Causally Explainable Decisions

Louis Anthony Cox
2021 Entropy  
For an AI agent to make trustworthy decision recommendations under uncertainty on behalf of human principals, it should be able to explain why its recommended decisions make preferred outcomes more likely  ...  The resulting causally explainable decisions make efficient use of available information to achieve goals in uncertain environments.  ...  Author Contributions : The author has read and agreed to the published version of the manuscript. Funding: This work was uncompensated and had no funding sources.  ... 
doi:10.3390/e23050601 pmid:34068183 fatcat:354opczpwba33l2xoaqu6thr34

Explaining Sentiment Classification with Synthetic Exemplars and Counter-Exemplars [chapter]

Orestis Lampridis, Riccardo Guidotti, Salvatore Ruggieri
2020 Lecture Notes in Computer Science  
The former are examples classified by the black box with the same label as the text to explain. The latter are examples classified with a different label (a form of counter-factuals).  ...  Both are close in meaning to the text to explain, and both are meaningful sentences – albeit they are synthetically generated. xspells generates neighbors of the text to explain in a latent space using  ...  The first author would like to thank Ioannis Mollas and Grigorios Tsoumakas for their support.  ... 
doi:10.1007/978-3-030-61527-7_24 fatcat:j5cum4gh65hkxdbnu7234fiosu

Explaining Link Prediction Systems based on Knowledge Graph Embeddings

Andrea Rossi, Donatella Firmani, Paolo Merialdo, Tommaso Teofili
2022 Proceedings of the 2022 International Conference on Management of Data  
In this paper, we propose the novel Kelpie explainability framework.  ...  Kelpie can be applied to any embedding-based LP models independently from their architecture, and it explains predictions by identifying the combinations of training facts that have enabled them.  ...  Alessandro Micarelli, Fabio Gasparetti and Alberto Paoluzzi for providing the computational resources for our work.  ... 
doi:10.1145/3514221.3517887 fatcat:g6sskzu7kncbbeyif7bpy7ec24

Rule Extraction in Unsupervised Anomaly Detection for Model Explainability: Application to OneClass SVM [article]

Alberto Barbado, Óscar Corcho, Richard Benjamins
2021 arXiv   pre-print
Together with that, we propose algorithms to compute metrics related with eXplainable Artificial Intelligence (XAI) regarding the "comprehensibility", "representativeness", "stability" and "diversity"  ...  We evaluate our proposals with different datasets, including real-world data coming from industry. With this, our proposal contributes to extend XAI techniques to unsupervised machine learning models.  ...  In order to optimize the exploration-exploitation of generating and evaluating data points, it uses a reinforcement learning approach with a Multi-Armed Bandit (MAB) approximation.  ... 
arXiv:1911.09315v5 fatcat:566dblz5frbmdl6ey4r5k4iexa

Explainable AI: A Review of Machine Learning Interpretability Methods

Pantelis Linardatos, Vasilis Papastefanopoulos, Sotiris Kotsiantis
2020 Entropy  
As a result, scientific interest in the field of Explainable Artificial Intelligence (XAI), a field that is concerned with the development of new methods that explain and interpret machine learning models  ...  Recent advances in artificial intelligence (AI) have led to its widespread industrial adoption, with machine learning systems demonstrating superhuman performance in a significant number of tasks.  ...  After highlighting the trend of the community to explore explainability only in terms of modelling, they proposed embracing explainability in other aspects of machine learning.  ... 
doi:10.3390/e23010018 pmid:33375658 pmcid:PMC7824368 fatcat:gv42gcovm5cxzl2kmdsluiegdi

Carousel Personalization in Music Streaming Apps with Contextual Bandits [article]

Walid Bendada and Guillaume Salha and Théo Bontempelli
2020 arXiv   pre-print
In this paper, we model carousel personalization as a contextual multi-armed bandit problem with multiple plays, cascade-based updates and delayed batch feedback.  ...  Media services providers, such as music streaming platforms, frequently leverage swipeable carousels to recommend personalized content to their users.  ...  Two versions, etc-seg-explore (n = 70) and etc-seg-exploit (n = 20) are evaluated. • ϵ-greedy-seg: recommends playlists randomly with probability ϵ, otherwise recommends the top-L with highest mean observed  ... 
arXiv:2009.06546v1 fatcat:kosq7gqqgvdujo3fusww6lvngq

Relational Boosted Bandits [article]

Ashutosh Kakadiya and Sriraam Natarajan and Balaraman Ravindran
2020 arXiv   pre-print
We empirically demonstrate the effectiveness and interpretability of RB2 on tasks such as link prediction, relational classification, and recommendations.  ...  RB2 enables us to learn interpretable and explainable models due to the more descriptive nature of the relational representation.  ...  the exploration-exploitation strategy F.  ... 
arXiv:2012.09220v1 fatcat:ektik4uoijcq7kcvtsqacquibm

From low-conflict polity to democratic civil peace: Explaining Zambian exceptionalism

Peter J. Burnell
2005 African Studies  
one primary "cause" will suffice to explain Zambian exceptionalism.  ...  This paper examines Zambia's experience against a background of general theories that try to explain conflict. It is an "interpretative case study".  ...  the country for personal gain.  ... 
doi:10.1080/00020180500355470 fatcat:ft73bkg2inelbnjmkvfkdf2yoq
« Previous Showing results 1 — 15 out of 2,981 results