24 Hits in 1.1 sec

Safe Exploration for Optimizing Contextual Bandits [article]

Rolf Jagerman and Ilya Markov and Maarten de Rijke
2020 arXiv   pre-print
Contextual bandit problems are a natural fit for many information retrieval tasks, such as learning to rank, text classification, recommendation, etc. However, existing learning methods for contextual bandit problems have one of two drawbacks: they either do not explore the space of all possible document rankings (i.e., actions) and, thus, may miss the optimal ranking, or they present suboptimal rankings to a user and, thus, may harm the user experience. We introduce a new learning method for
more » ... ntextual bandit problems, Safe Exploration Algorithm (SEA), which overcomes the above drawbacks. SEA starts by using a baseline (or production) ranking system (i.e., policy), which does not harm the user experience and, thus, is safe to execute, but has suboptimal performance and, thus, needs to be improved. Then SEA uses counterfactual learning to learn a new policy based on the behavior of the baseline policy. SEA also uses high-confidence off-policy evaluation to estimate the performance of the newly learned policy. Once the performance of the newly learned policy is at least as good as the performance of the baseline policy, SEA starts using the new policy to execute new actions, allowing it to actively explore favorable regions of the action space. This way, SEA never performs worse than the baseline policy and, thus, does not harm the user experience, while still exploring the action space and, thus, being able to find an optimal policy. Our experiments using text classification and document retrieval confirm the above by comparing SEA (and a boundless variant called BSEA) to online and offline learning methods for contextual bandit problems.
arXiv:2002.00467v1 fatcat:2fbpifkvhvdo3m5ko5qrlf42z4

A Directional Diffusion Algorithm for Inpainting [article]

Jan Deriu, Rolf Jagerman, Kai-En Tsay
2015 arXiv   pre-print
The problem of inpainting involves reconstructing the missing areas of an image. Inpainting has many applications, such as reconstructing old damaged photographs or removing obfuscations from images. In this paper we present the directional diffusion algorithm for inpainting. Typical diffusion algorithms are bad at propagating edges from the image into the unknown masked regions. The directional diffusion algorithm improves on the regular diffusion algorithm by reconstructing edges more
more » ... ly. It scores better than regular diffusion when reconstructing images that are obfuscated by a text mask.
arXiv:1511.03464v1 fatcat:6rgwabvbirbd3ccbuagj3yav7i

Modeling Label Ambiguity for Neural List-Wise Learning to Rank [article]

Rolf Jagerman, Julia Kiseleva, Maarten de Rijke
2017 arXiv   pre-print
List-wise learning to rank methods are considered to be the state-of-the-art. One of the major problems with these methods is that the ambiguous nature of relevance labels in learning to rank data is ignored. Ambiguity of relevance labels refers to the phenomenon that multiple documents may be assigned the same relevance label for a given query, so that no preference order should be learned for those documents. In this paper we propose a novel sampling technique for computing a list-wise loss
more » ... at can take into account this ambiguity. We show the effectiveness of the proposed method by training a 3-layer deep neural network. We compare our new loss function to two strong baselines: ListNet and ListMLE. We show that our method generalizes better and significantly outperforms other methods on the validation and test sets.
arXiv:1707.07493v1 fatcat:fbxyo3ooajdbtee2gba2rrmj34

The fifteen year struggle of decentralizing privacy-enhancing technology [article]

Rolf Jagerman, Wendo Sabée, Laurens Versluis, Martijn de Vos, Johan Pouwelse
2014 arXiv   pre-print
Ever since the introduction of the internet, it has been void of any privacy. The majority of internet traffic currently is and always has been unencrypted. A number of anonymous communication overlay networks exist whose aim it is to provide privacy to its users. However, due to the nature of the internet, there is major difficulty in getting these networks to become both decentralized and anonymous. We list reasons for having anonymous networks, discern the problems in achieving
more » ... on and sum up the biggest initiatives in the field and their current status. To do so, we use one exemplary network, the Tor network. We explain how Tor works, what vulnerabilities this network currently has, and possible attacks that could be used to violate privacy and anonymity. The Tor network is used as a key comparison network in the main part of the report: a tabular overview of the major anonymous networking technologies in use today.
arXiv:1404.4818v1 fatcat:7yfg6ske2fhybodp665gtkflny

Query-Level Ranker Specialization

Rolf Jagerman, Harrie Oosterhuis, Maarten de Rijke
2017 International Conference on the Theory of Information Retrieval  
Traditional Learning to Rank models optimize a single ranking function for all available queries. is assumes that all queries come from a homogenous source. Instead, it seems reasonable to assume that queries originate from heterogenous sources, where certain queries may require documents to be ranked di erently. We introduce the Specialized Ranker Model which assigns queries to di erent rankers that become specialized on a subset of the available queries. We provide a theoretical foundation
more » ... this model starting from the listwise Placke -Luce ranking model and derive a computationally feasible expectation-maximization procedure to infer the model's parameters. Furthermore we experiment using a noisy oracle to model the risk/reward tradeo that exists when deciding which specialized ranker to use for unseen queries.
dblp:conf/ictir/JagermanOR17 fatcat:4443veonhjbv3evnbobacbneqa

Unbiased Learning to Rank: Counterfactual and Online Approaches

Harrie Oosterhuis, Rolf Jagerman, Maarten de Rijke
2020 Companion Proceedings of the Web Conference 2020  
This tutorial is about Unbiased Learning to Rank, a recent research field that aims to learn unbiased user preferences from biased user interactions. We will provide an overview of the two main families of methods in Unbiased Learning to Rank: Counterfactual Learning to Rank (CLTR) and Online Learning to Rank (OLTR) and their underlying theory. First, the tutorial will start with a brief introduction to the general Learning to Rank (LTR) field and the difficulties user interactions pose for
more » ... itional supervised LTR methods. The second part will cover Counterfactual Learning to Rank (CLTR), a LTR field that sprung out of click models. Using an explicit model of user biases, CLTR methods correct for them in their learning process and can learn from historical data. Besides these methods, we will also cover practical considerations, such as how certain biases can be estimated. In the third part of the tutorial we focus on Online Learning to Rank (OLTR), methods that learn by directly interacting with users and dealing with biases by adding stochasticity to displayed results. We will cover cascading bandits, dueling bandit techniques and the most recent pairwise differentiable approach. Finally, in the concluding part of the tutorial, both approaches are contrasted, highlighting their relative strengths and weaknesses, and presenting future directions of research. For LTR practitioners our comparison gives guidance on how the choice between methods should be made. For the field of Information Retrieval (IR) we aim to provide an essential guide on unbiased LTR to understanding and choosing between methodologies.
doi:10.1145/3366424.3383107 dblp:conf/www/OosterhuisJR20 fatcat:lbuptnrrvbapllvdegqygtlkqm

Computing Web-scale Topic Models using an Asynchronous Parameter Server

Rolf Jagerman, Carsten Eickhoff, Maarten de Rijke
2017 Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '17  
Topic models such as Latent Dirichlet Allocation (LDA) have been widely used in information retrieval for tasks ranging from smoothing and feedback methods to tools for exploratory search and discovery. However, classical methods for inferring topic models do not scale up to the massive size of today's publicly available Web-scale data sets. The state-of-the-art approaches rely on custom strategies, implementations and hardware to facilitate their asynchronous, communication-intensive
more » ... We present APS-LDA, which integrates state-of-the-art topic modeling with cluster computing frameworks such as Spark using a novel asynchronous parameter server. Advantages of this integration include convenient usage of existing data processing pipelines and eliminating the need for disk writes as data can be kept in memory from start to finish. Our goal is not to outperform highly customized implementations, but to propose a general high-performance topic modeling framework that can easily be used in today's data processing pipelines. We compare APS-LDA to the existing Spark LDA implementations and show that our system can, on a 480-core cluster, process up to 135 times more data and 10 times more topics without sacrificing model quality.
doi:10.1145/3077136.3084135 dblp:conf/sigir/JagermanER17 fatcat:acilgidzindmljvhkqtgkv6fl4

Accelerated Convergence for Counterfactual Learning to Rank [article]

Rolf Jagerman, Maarten de Rijke
2020 pre-print
Counterfactual Learning to Rank (LTR) algorithms learn a ranking model from logged user interactions, often collected using a production system. Employing such an offline learning approach has many benefits compared to an online one, but it is challenging as user feedback often contains high levels of bias. Unbiased LTR uses Inverse Propensity Scoring (IPS) to enable unbiased learning from logged user interactions. One of the major difficulties in applying Stochastic Gradient Descent (SGD)
more » ... aches to counterfactual learning problems is the large variance introduced by the propensity weights. In this paper we show that the convergence rate of SGD approaches with IPS-weighted gradients suffers from the large variance introduced by the IPS weights: convergence is slow, especially when there are large IPS weights. To overcome this limitation, we propose a novel learning algorithm, called CounterSample, that has provably better convergence than standard IPS-weighted gradient descent methods. We prove that CounterSample converges faster and complement our theoretical findings with empirical results by performing extensive experimentation in a number of biased LTR scenarios -- across optimizers, batch sizes, and different degrees of position bias.
doi:10.1145/3397271.3401069 arXiv:2005.10615v1 fatcat:rjgwhs27tjcxxjqcyb2ti3cbei

Page 1036 of Mathematical Reviews Vol. 42, Issue 4 [page]

1971 Mathematical Reviews  
T. ......00000. 5532 | Jagerman, D. ..........- 5698 Ti se udbobe ce eee 4905 | Gorgalean, I.  ...  A. ......... 5671 Grigorieff, Rolf Dieter .... 5456 | Hiramatu, Hitosi ........ 5193 Elliott, George A. ........ 5068 | Gandz, Solomon ........ *4358 | Grillet, Mireille P. ....... 4597 | Hironaka, Heisuke  ... 

Page 1857 of Mathematical Reviews Vol. , Issue 89C [page]

1989 Mathematical Reviews  
SEs. enaadininaannswtnubinnonmniiaiats 58037 Jagerman, D. ...... Jahn, J. Jahn, Johannes Jahn, K.-U. Jahnke, Hans Niels Jain, M. K. .....  ...  See Jitina, Miloslav Jifina, Miloslav Jédar, Lucas I itniithniitntinindvninaitinlizas ‘ Johansson, Rolf IE ecinenantnctncsoscnnaiiin 73004 Johnson, Charles R. 15005,15012 Johnson, Claes Johnstone, D.  ... 

Page 1264 of Mathematical Reviews Vol. , Issue 84c [page]

1984 Mathematical Reviews  
Jarvie, J.P. ..... 4 Jaworski, M. .. 35097, 58047 > re 06006 Jeffries, Carson Jeltsch, Rolf Jennings, R.  ...  Isermann, Rolf Ishihara, T. See Ishihara, Téru Ishihara, Téru Ishii, Hiroaki Ishimori, Yuji IshlinskiY, A. Yu. Isidori, Alberto Iskenderov, B.  ... 

Page 4395 of Mathematical Reviews Vol. , Issue 86i [page]

1986 Mathematical Reviews  
See «81004 Jagerman, D. L. ...... 60233 Jain, Naresh C. ....... 60090 Jain, U. C. Ja’Ja’, J. Jakébczak, Piotr Semmes, A.  ...  Jeltsch, Rolf Jenkins, James A. .... Jennen, Christel Jennings, R. E. ....... Jennison, C. Jensen, D. R. .. Jensen, E. B. Jensen, Helge Elbrgnd Jepson, Allan D. 58018,65022 Jeroslow, R.  ... 

Page 4675 of Mathematical Reviews Vol. , Issue 93h [page]

1993 Mathematical Reviews  
Klonecki, Witold Klétzler, Rolf I Ole eens ctdyesesccccnacss 49058 Knauf, Andreas Kneip, Alois Knessl, Charles Knezevic¢, Julka Knezhevich-Milyanovich, Yu.  ...  swnncacidicsades 17006 Jacobs, Konrad Jacqmin, David Jacquemin, Michel Jacquotte, Olivier-Pierre Jaeger, Francois Jaffard, Stéphane Jaffe, A. 28011,42024,47035 Rusdabescenescacdesiceeces See « 81002 Jager, Willi Jagerman  ... 

Page 5073 of Mathematical Reviews Vol. , Issue 95h [page]

1995 Mathematical Reviews  
See Hiinlich, Rolf II os 0520632 canucenunsen 35114 et ES.  ...  E 90078 Jacobson, Michael S 05093 Jacod, J 62151 Jacquin, Laurent 76064 Jacroux, Mike 62129 Jaeger, Francois .. 82008 Jafari, F. 32028,41035 Jaffard, Stéphane 62154 Jagadeesan, Radha 03124 Jagerman, David  ... 

Page 1245 of Mathematical Reviews Vol. , Issue 95b [page]

1995 Mathematical Reviews  
, Fe COE... . cee cesessceeveces 94020 Fe BS ivincsynssccveescannntead 22018 SED cn.onncinoatespaecseestess 35081 MOOI, Si.ncescgsecesscavancneees 57011 BU ED ck sccconkecnees wees 28007 Johannesson, Rolf  ...  sisccincessasssnscesacee 17002 os, A rar See 93061 2 ee 81017 Se crn ernes 68035 _ ___ i aR SES ES 42029 Jaffe, Arthur ........ 00004, See also 00003 See Jagannathan, Ramaswamy Jagannathan, Ramaswamy ............ 81087 Jagerman  ... 
« Previous Showing results 1 — 15 out of 24 results