Filters








725,074 Hits in 3.7 sec

K-Space Interactive Search

Peter Wilkins, Alan F. Smeaton, Noel E. O'Connor, Daragh Byrne
2008 Proceedings of the 2008 international conference on Content-based image and video retrieval - CIVR '08  
In this paper we will present the K-Space 1 Interactive Search system for content-based video information retrieval to be demonstrated in the VideOlympics.  ...  Acknowledgments The research leading to this paper was supported by the European Commission under contract FP6-027026 (K-Space).  ...  K-Space is a European Network of Excellence (NoE) in semantic inference for semi-automatic annotation and retrieval of multimedia content (http://www.k-space.eu)  ... 
doi:10.1145/1386352.1386428 dblp:conf/civr/WilkinsSOB08 fatcat:dziqipbo6ncpfniwvuhdidoohu

A Data-Driven Interactive System for Aerodynamic and User-centred Generative Vehicle Design

Muhammad Usama, Aqib Arif, Farhan Haris, Shahroz Khan, S. Kamran Afaq, Shahrukh Rashid
2021 2021 International Conference on Artificial Intelligence (ICAI)  
To avoid the curse of dimensionality, the qualitative search process first builds up a lower-dimensional representation of a given design space, which is then explored using the unsupervised k-means clustering  ...  The quantitative search process explores the design space to find an optimal design in terms of performance criterion such as drag coefficient.  ...  x l m,k = x u m,k + κ T T ln x t,k x l m,k , x l m,k = x u m,k + κ T T ln x t,k x l m,k , ∀k ∈ {1, 2, . . . , n} (5) Here, κ T is the shrink rate initialised by the user in the T th interaction and ranges  ... 
doi:10.1109/icai52203.2021.9445243 fatcat:t3mvwfdltzblpmmq23zmll4iji

AutoML for Deep Recommender Systems: A Survey [article]

Ruiqi Zheng, Liang Qu, Bin Cui, Yuhui Shi, Hongzhi Yin
2022 arXiv   pre-print
Secondly, we present a taxonomy as a classification framework containing embedding dimension search, feature interaction search, model design search and other components search.  ...  Furthermore, we put a particular emphasis on the search space and search strategy, as they are the common thread to connect all methods within each category and enable practitioners to analyze and compare  ...  The search space is 𝑂 (2 𝑓 𝐾 ), where pre-defined 𝐾 represents the maximum order of feature interactions. Firstly, for 𝑝 𝑡ℎ order of feature interactions, it selects 𝑛 𝑝 feature sets.  ... 
arXiv:2203.13922v2 fatcat:awvhxzkopfcafgryusmjl4davu

Efficient Neural Interaction Function Search for Collaborative Filtering [article]

Quanming Yao, Xiangning Chen, James Kwok, Yong Li, Cho-Jui Hsieh
2020 arXiv   pre-print
By examining and generalizing existing CF approaches, an expressive SIF search space is designed and represented as a structured multi-layer perceptron.  ...  Motivated by the recent success of automated machine learning (AutoML), we propose in this paper the search for simple neural interaction functions (SIF) in CF.  ...  ] AutoML SIF (proposed) searched O((m + n)k) O(k) -- In this paper, motivated by the success of automated machine learning (AutoML) [20, 44] , we consider formulating the search for interaction functions  ... 
arXiv:1906.12091v3 fatcat:xnwqonn4eva5rnvd54qg7rpl2i

NAS-CTR

Guanghui Zhu, Feng Cheng, Defu Lian, Chunfeng Yuan, Yihua Huang
2022 Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval  
Unfortunately, given a domain-specific dataset, searching effective feature interaction operations and combinations from a huge candidate space requires significant expert experience and computational  ...  First, we design a novel and expressive architecture search space and a continuous relaxation scheme to make the search space differentiable.  ...  In the future, we plan to introduce more interaction operations to the search space.  ... 
doi:10.1145/3477495.3532030 fatcat:wgomfzjfhfd65kdully3wtsoze

Faceted search, social networking and interactive semantics

Hai Zhuge, Yorick Wilks
2013 World wide web (Bussum)  
Multi-dimensional classifications, called Resource Space Model, have been used to manage various resources [13] [14] [15] [16] .  ...  A faceted search system can display different facets of the same set of Web pages at different steps in a search.  ...  Figure 1 1 Search, interaction and social networking in a complex space  ... 
doi:10.1007/s11280-013-0216-6 fatcat:kx3czhsp5zh3bnmih2uq4bzpra

Planning with Selective Physics-based Simulation for Manipulation Among Movable Objects [article]

Muhammad Suhail Saleem, Maxim Likhachev
2020 arXiv   pre-print
Use of physics-based simulation as a planning model enables a planner to reason and generate plans that involve non-trivial interactions with the world.  ...  A physics-based simulator allows a planner to reason about the effects of interactions with these objects and to generate a plan that grasps the milk container successfully.  ...  Let S sel represent the search space for our approach and π sel be the plan generated by a search in this space.  ... 
arXiv:2003.06743v1 fatcat:mjpc7plhwfbhzitgqzkcq357oa

Deep Sequential Recommendation for Personalized Adaptive User Interfaces

Harold Soh, Scott Sanner, Madeleine White, Greg Jamieson
2017 Proceedings of the 22nd International Conference on Intelligent User Interfaces - IUI '17  
Specifically, interaction histories are embedded in a learned space along with users and interaction elements; this allows the AUI to query and recommend likely next actions based on similar usage patterns  ...  But this is difficult to achieve with many interaction elements when data-per-user is sparse.  ...  In this respect, the Euclidean space is preferable to dot-product spaces; the search for preferred items reduces to a k-nearest-neighbors (k-NN) search since a u,t+1,j = x u − x j 2 2 + h u,t − x j 2 2  ... 
doi:10.1145/3025171.3025207 dblp:conf/iui/SohSWJ17 fatcat:2dgheq2iizd2jdsv3ddqj6yu3m

The Role of High-Dimensional Diffusive Search, Stabilization, and Frustration in Protein Folding

Supreecha Rimratchada, Tom C.B. McLeish, Sheena E. Radford, Emanuele Paci
2014 Biophysical Journal  
Proteins explore the conformation space by searching these flat subspaces to find a narrow energetic alley that we call a hypergutter and then explore the next, lower-dimensional, subspace.  ...  Such a framework provides an effective representation of the energy landscape and folding kinetics that does justice to the essential characteristic of high-dimensionality of the search-space.  ...  The intermediate states can be stabilized by both native and nonnative interactions and nonnative interactions can accelerate protein folding by forming intermediates, which partition the search space.  ... 
doi:10.1016/j.bpj.2014.01.051 pmid:24739172 pmcid:PMC4008833 fatcat:yo3vxc76rneyzmd7ynbjfxwcia

Protein Folding in High-Dimensional Spaces: Hypergutters and the Role of Nonnative Interactions

T.C.B. McLeish
2005 Biophysical Journal  
searches on the folding pathway, as well as native interactions; 3), sequential barriers will commonly be found, even in "two-state" proteins; 4), very early times will show characteristic departures  ...  It suggests that kinetic strategies for fast folding may be encoded rather generally in nonnative as well as in native interactions.  ...  Radford, K. Plaxco, J.-E. Shea, M. Oliveberg, and D.  ... 
doi:10.1529/biophysj.103.036616 pmid:15501939 pmcid:PMC1304996 fatcat:ppkhibxjarbvhh75j6uxfdclei

Interactive Markov Models of Optimization Search Strategies

Haiping Ma, Dan Simon, Minrui Fei, Hongwei Mo
2017 IEEE Transactions on Systems, Man & Cybernetics. Systems  
The probability of selection is linearly proportional to the number of individuals at each point of the search space.  ...  We exactly model these optimization search strategies with interactive Markov models. We present simulation results to confirm the interactive Markov model theory.  ...  individuals from the search space (recall that K is the cardinality of the search space, and {x i : i = 1, . . . , K} is the search space).  ... 
doi:10.1109/tsmc.2015.2507588 fatcat:diqjohzkezhotjkssvl2uzt2hq

Shopping in the Multiverse: A Counterfactual Approach to In-Session Attribution [article]

Jacopo Tagliabue, Bingqing Yu
2020 arXiv   pre-print
induced by prod2vec embeddings; we show how natural language queries can be effectively represented in the same space and how "search intervention" can be performed to assess causal contribution.  ...  We tackle the challenge of in-session attribution for on-site search engines in eCommerce.  ...  after the search interaction (w 1 ), green is the timeline generated before the search interaction (w 2 ). • C(CU ) = k; • C(SE) = k * r where 0.0 ≤ r ≤ 1.0; • C(ME) = k * r /2; • C(SR) = k * r /4.  ... 
arXiv:2007.10087v1 fatcat:nzptlux2anh55cwuw7jkj7hzfu

The unitary dynamical state-locking process, the HSSS quantum search process, and the quantum-computing speedup theory [article]

Xijia Miao
2016 arXiv   pre-print
The fundamental interaction mentioned above and the fundamental quantum-computing resource play the main role in speeding up the search process.  ...  That it can change the quantum-state difference is original from the fundamental interaction between the quantum physical laws (i.e., the unitary quantum dynamics and the Hilbert-space symmetric structure  ...  Its quantum-computing speedup is original from the interaction mentioned above and hence the interaction is also responsible for the search-space dynamical reduction.  ... 
arXiv:1612.05969v2 fatcat:7hblaawybbfynjqrslmrop6dbu

Minimal Interaction Search in Recommender Systems

Branislav Kveton, Shlomo Berkovsky
2015 Proceedings of the 20th International Conference on Intelligent User Interfaces - IUI '15  
We propose generalized linear search (GLS), a combination of linear and generalized searches that brings together the benefits of both approaches.  ...  In this work, we look into user interaction with the recommendation list, aiming to devise a method that allows users to discover items of interest in a minimal number of interactions.  ...  First, the length of interaction decreases with K, as larger values of K allow for a fine-grained partitioning of the version space. Second, for all values of K, GLS significantly outperforms GS.  ... 
doi:10.1145/2678025.2701367 dblp:conf/iui/KvetonB15 fatcat:zjghenv77vdmdbp5o7weua2o6u

Protein folding in high-dimensional spaces:hypergutters and the role of non-native interactions [article]

T.C.B. McLeish
2003 arXiv   pre-print
; (5) contributions of non-native interactions to phi-values are calculable, and may be significant.  ...  searches on the folding pathway; (3) sequential barriers will commonly be found, even in "two-state"proteins; (4) very early times will show charactreristic departures from single-exponential kinetics  ...  factors): k f = ∆ 1 + ∆ k 1+2 + 1 1 + ∆ k 3 (8) where k 1+2 = τ −1 1+2 and the slow rate k 3 = τ −1 3 ≈ σk 2 .  ... 
arXiv:physics/0311028v2 fatcat:m6vaivchnvfizg4ywsykt7vyk4
« Previous Showing results 1 — 15 out of 725,074 results