Bayesian Optimization for Optimizing Retrieval Systems

Li, Dan; Kanoulas, Evangelos

doi:10.1145/3159652.3159665

Cited by 12 publications

(14 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…BO is a sequential search strategy for the global optimization of an expensive black-box function f (x) (Mockus 2012). It first emerged as a successful strategy in many machine learning applications (Bergstra et al 2011, Snoek et al 2012, Bergstra et al 2013, Swersky et al 2013, 2014, Yogatama and Smith 2015, and has lately been employed in many other areas including robotics (Lizotte et al 2007, Calandra et al 2016, sensor networks (Garnett et al 2010), environmental monitoring (Marchant and Ramos 2012), information extraction and retrieval (Wang et al 2014, Li andKanoulas 2018), and game theory (Picheny et al 2016).…”

Section: System Optimization Using Bayesian Optimizationmentioning

confidence: 99%

A framework to integrate mode choice in the design of mobility-on-demand systems

Liu

Bansal

Samaranayake

2019

Transportation Research Part C: Emerging Technologies

View full text Add to dashboard Cite

Mobility-on-Demand (MoD) systems are generally designed and analyzed for a fixed and exogenous demand, but such frameworks fail to answer questions about the impact of these services on the urban transportation system, such as the effect of induced demand and the implications for transit ridership. In this study, we propose a unified framework to design, optimize and analyze MoD operations within a multimodal transportation system where the demand for a travel mode is a function of its level of service. An application of Bayesian optimization (BO) to derive the optimal supply-side MoD parameters (e.g., fleet size and fare) is also illustrated. The proposed framework is calibrated using the taxi demand data in Manhattan, New York. Travel demand is served by public transit and MoD services of varying passenger capacities (1, 4 and 10), and passengers are predicted to choose travel modes according to a mode choice model. This choice model is estimated using stated preference data collected in New York City. The convergence of the multimodal supply-demand system and the superiority of the BO-based optimization method over earlier approaches are established through numerical experiments. We finally consider a policy intervention where the government imposes a tax on the ride-hailing service and illustrate how the proposed framework can quantify the pros and cons of such policies for different stakeholders.

show abstract

Section: System Optimization Using Bayesian Optimizationmentioning

confidence: 99%

A framework to integrate mode choice in the design of mobility-on-demand systems

Liu

Bansal

Samaranayake

2019

Transportation Research Part C: Emerging Technologies

View full text Add to dashboard Cite

show abstract

“…They proposed a technique that can combine multiple terms using fuzzy logic. Also, Bayesian [40] with many retrieval models can effectively improve search performance. They considered using models that can allow users to search for information using a query and optimize the results.…”

Section: B Query Expansion Methods Based On Relevance Feedbacksmentioning

confidence: 99%

Query Expansion based on Explicit-Relevant Feedback and Synonyms for English Quran Translation Information Retrieval

Yusuf¹,

Yunus²,

Wahid³

2019

IJACSA

View full text Add to dashboard Cite

Search engines are commonly present as information retrieval applications that help to retrieve relevant information from different domain areas. The crucial part of improving the quality of search engine is based on query expansion, which expands the query with additional information to match additional important documents. This paper presents a query expansion approach that utilizes explicit relevant feedback with word synonyms and semantic relatedness. We describe the possibility and demonstrations based on the experimental work pertain to search engines where relevant judgment and word synonyms can improve search quality. In order to show the level of improving the proposed approach, we compared the results obtained from the experiments based on Yusuf Ali, Arberry and Sarwar Quran datasets. The proposed approach shows improvement over other methods.

show abstract

“…For these approaches, given an unknown function f , we are allowed to evaluate f (x) for any value x ∈ X in an effort to maximize (or minimize) f (x). Typical approaches include Bayesian optimization [11,17] as well as hybrid methods using random search alongside multi-armed bandit techniques [19]. In particular, [17] discusses how Bayesian optimization can be used for optimizing retrieval systems.…”

Section: Related Workmentioning

confidence: 99%

“…Typical approaches include Bayesian optimization [11,17] as well as hybrid methods using random search alongside multi-armed bandit techniques [19]. In particular, [17] discusses how Bayesian optimization can be used for optimizing retrieval systems. These approaches generally assume some prior belief over f and draw samples to get a posterior that better approximates f .…”

Section: Related Workmentioning

confidence: 99%

“…Given that an industrial search system can be very complex, a natural and common approach for parameter tuning is to treat the entire system as a black box and run A/B experiments with either some sort of parameter sweep (e.g. grid search, coordinate ascent) [22] or black-box optimization [17] over various parameters in a guess-and-check manner. The drawback to this is that 1) the user experience can be degraded due to exposure to poor quality results, 2) separate online experiments are required for each test set of parameters, and 3) the tuning process can take months.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Parameter Tuning in Personal Search Systems

Chen

Wang

Qin

et al. 2020

Proceedings of the 13th International Conference on Web Search and Data Mining

View full text Add to dashboard Cite

The effectiveness of information retrieval systems is heavily dependent on how various parameters are tuned. One option to find these parameters is to run multiple online experiments using a parameter sweep approach in order to optimize the search system. There are multiple downsides of this approach, including the fact that it may lead to a poor experience for users. Another option is to do offline evaluation, which can act as a safeguard against potential quality issues. Offline evaluation requires a validation set of data that can be benchmarked against different parameter settings. However, for search over personal corpora, e.g. email and file search, it is impractical and often impossible to get a complete representative validation set due to the inability to save raw queries and document information. In this work, we show how to do offline parameter tuning with only a partial validation set. In addition, we demonstrate how to do parameter tuning in situations when we have complete knowledge of the internal implementation of the search system (white-box tuning), as well as situations where we have only partial knowledge (grey-box tuning). The resulting method provides a way of performing offline parameter tuning in a privacy-preserving manner. We demonstrate the effectiveness of the proposed approach by reporting the results from search ranking experiments performed on two large-scale personal search systems. CCS CONCEPTS• Information systems → Evaluation of retrieval results.

show abstract

Bayesian Optimization for Optimizing Retrieval Systems

Cited by 12 publications

References 23 publications

A framework to integrate mode choice in the design of mobility-on-demand systems

A framework to integrate mode choice in the design of mobility-on-demand systems

Query Expansion based on Explicit-Relevant Feedback and Synonyms for English Quran Translation Information Retrieval

Parameter Tuning in Personal Search Systems

Contact Info

Product

Resources

About