scholarly journals Linear Bandits with Feature Feedback

2020 ◽  
Vol 34 (04) ◽  
pp. 5331-5338
Author(s):  
Urvashi Oswal ◽  
Aniruddha Bhargava ◽  
Robert Nowak
Keyword(s):  
Computational Complexity ◽  
Prior Knowledge ◽  
Real World ◽  
Time Horizon ◽  
The Novel ◽  
Bandit Problem ◽  
Real World Applications ◽  
New Form ◽  
Feature Feedback ◽  
Over Time

This paper explores a new form of the linear bandit problem in which the algorithm receives the usual stochastic rewards as well as stochastic feedback about which features are relevant to the rewards, the latter feedback being the novel aspect. The focus of this paper is the development of new theory and algorithms for linear bandits with feature feedback which can achieve regret over time horizon T that scales like k√T, without prior knowledge of which features are relevant nor the number k of relevant features. In comparison, the regret of traditional linear bandits is d√T, where d is the total number of (relevant and irrelevant) features, so the improvement can be dramatic if k ≪ d. The computational complexity of the algorithm is proportional to k rather than d, making it much more suitable for real-world applications compared to traditional linear bandits. We demonstrate the performance of the algorithm with synthetic and real human-labeled data.

scholarly journals Design and implementation of aggregate functions in the DLV system

2008 ◽  
Vol 8 (5-6) ◽  
pp. 545-580 ◽  
Author(s):  
WOLFGANG FABER ◽  
GERALD PFEIFER ◽  
NICOLA LEONE ◽  
TINA DELL'ARMI ◽  
GIUSEPPE IELPA
Keyword(s):  
Computational Complexity ◽  
Logic Programming ◽  
Real World ◽  
State Of The Art ◽  
Complexity Class ◽  
Natural Manner ◽  
Design And Implementation ◽  
Finite Structures ◽  
Real World Applications ◽  
Aggregate Functions

AbstractDisjunctive logic programming (DLP) is a very expressive formalism. It allows for expressing every property of finite structures that is decidable in the complexity class ΣP2(=NPNP). Despite this high expressiveness, there are some simple properties, often arising in real-world applications, which cannot be encoded in a simple and natural manner. Especially properties that require the use of arithmetic operators (like sum, times, or count) on a set or multiset of elements, which satisfy some conditions, cannot be naturally expressed in classic DLP. To overcome this deficiency, we extend DLP by aggregate functions in a conservative way. In particular, we avoid the introduction of constructs with disputed semantics, by requiring aggregates to be stratified. We formally define the semantics of the extended language (called ), and illustrate how it can be profitably used for representing knowledge. Furthermore, we analyze the computational complexity of , showing that the addition of aggregates does not bring a higher cost in that respect. Finally, we provide an implementation of in DLV—a state-of-the-art DLP system—and report on experiments which confirm the usefulness of the proposed extension also for the efficiency of computation.

scholarly journals Cardinality Estimators do not Preserve Privacy

2019 ◽  
Vol 2019 (2) ◽  
pp. 26-46 ◽  
Author(s):  
Damien Desfontaines ◽  
Andreas Lochbihler ◽  
David Basin
Keyword(s):  
Prior Knowledge ◽  
Real World ◽  
Private Information ◽  
Differential Privacy ◽  
Risk Mitigation ◽  
Mitigation Strategies ◽  
Raw Data ◽  
Real World Applications ◽  
Risk Mitigation Strategies ◽  
Abstract Notion

Abstract Cardinality estimators like HyperLogLog are sketching algorithms that estimate the number of distinct elements in a large multiset. Their use in privacy-sensitive contexts raises the question of whether they leak private information. In particular, can they provide any privacy guarantees while preserving their strong aggregation properties? We formulate an abstract notion of cardinality estimators, that captures this aggregation requirement: one can merge sketches without losing precision. We propose an attacker model and a corresponding privacy definition, strictly weaker than differential privacy: we assume that the attacker has no prior knowledge of the data. We then show that if a cardinality estimator satisfies this definition, then it cannot have a reasonable level of accuracy. We prove similar results for weaker versions of our definition, a nd a nalyze t he p rivacy o f existing algorithms, showing that their average privacy loss is significant, e ven f or m ultisets w ith l arge cardinalities. We conclude that strong aggregation requirements are incompatible with any reasonable definition o f privacy, and that cardinality estimators should be considered as sensitive as raw data. We also propose risk mitigation strategies for their real-world applications.

AN EVALUATION OF THE NOVEL "NORMALIZED-REDUNDANCY" REPRESENTATION FOR PLANAR CURVES

1996 ◽  
Vol 10 (07) ◽  
pp. 769-789 ◽  
Author(s):  
J. FDEZ-VALDIVIA ◽  
J. A. GARCIA ◽  
M. GARCIA-SILVENTE
Keyword(s):  
Computational Complexity ◽  
Prior Knowledge ◽  
Shape Representation ◽  
Shape Recognition ◽  
General Purpose ◽  
The Novel ◽  
Time And Space ◽  
Novel Approach ◽  
Order Polynomial ◽  
Different Levels

In this paper, a novel approach for improving model-based recognition is proposed. Our approach provides a suitable shape representation by extracting only the most significant scales that best describe a planar noisy curve. The proposed representation satisfies several necessary criteria for general-purpose shape representation methods. The representation is capable of dealing with different levels of noise, it does not require user-set parameters or prior knowledge about the curve's nature, it also has a very low-order polynomial computational complexity in time and space. Hence such a shape representation is very useful for shape recognition. The method depends on the connection between the redundancy of two signals' smoothed versions and the essential structure being simultaneously isolated in both versions. Two different ways of formulating this approach are described in this paper: the global "normalized-redundancy" representation and the local "normalized-redundancy" representation. Results of applying the proposed formulation to synthetic and real 2-D shapes are presented.

Generalization analysis of multi-modal metric learning

2016 ◽  
Vol 14 (04) ◽  
pp. 503-521
Author(s):  
Yunwen Lei ◽  
Yiming Ying
Keyword(s):  
Prior Knowledge ◽  
Real World ◽  
Learning Algorithms ◽  
Metric Learning ◽  
Modal Data ◽  
Rademacher Complexity ◽  
Generalization Bounds ◽  
Complexity Result ◽  
Real World Applications

Multi-modal metric learning has recently received considerable attention since many real-world applications involve multi-modal data. However, there is relatively little study on the generalization analysis of the associated learning algorithms. In this paper, we bridge this theoretical gap by deriving its generalization bounds using Rademacher complexities. In particular, we establish a general Rademacher complexity result by systematically analyzing the behavior of the resulting models with various regularizers, e.g., [Formula: see text]-regularizer on the modality level with either a mixed [Formula: see text]-norm or a Schatten norm on each modality. Our results and the discussion followed help to understand how the prior knowledge can be exploited by selecting an appropriate regularizer.

ONTOLOGY-BASED INFORMATION EXTRACTION FROM PDF DOCUMENTS WITH XONTO

2009 ◽  
Vol 18 (05) ◽  
pp. 673-695 ◽  
Author(s):  
ERMELINDA ORO ◽  
MASSIMO RUFFOLO ◽  
DOMENICO SACCÀ
Keyword(s):  
Information Extraction ◽  
Real World ◽  
Portable Document Format ◽  
The Novel ◽  
Tabular Form ◽  
Military Intelligence ◽  
Real World Applications ◽  
Document Format

Information extraction is of paramount importance in several real world applications in the areas of business, competitive and military intelligence because it enables to acquire information contained in unstructured documents and store them in structured forms. Unstructured documents have different internal encodings, one of the most diffused encoding is the visualization-oriented Adobe portable document format (PDF). Although several sophisticated and indeed complex approaches were proposed, they are still limited in many aspects. In particular, existing information extraction systems cannot be applied to PDF documents because of their completely unstructured nature that pose many issues in defining IE approaches. In this paper the novel ontology-based system named XONTO, that allows the semantic extraction of information from PDF documents, is presented. The XONTO system is founded on the idea of self-describing ontologies in which objects and classes can be equipped by a set of rules named descriptors. These rules represent patterns that allow to automatically recognize and extract ontology objects contained in PDF documents also when information is arranged in tabular form. This way a self-describing ontology expresses the semantic of the information to extract and the rules that, in turn, populate itself. In the paper XONTO system behaviors and structure are sketched by means of a running example.

Religion and Memory in Tacitus' Annals

2018 ◽  
Author(s):  
Kelly E. Shannon-Henderson
Keyword(s):  
Cultural Memory ◽  
Family Members ◽  
Religious Life ◽  
Memory Theory ◽  
State Changes ◽  
New Form ◽  
Religious Material ◽  
Over Time

This study demonstrates the importance of references to religious material in Tacitus’ Annals by analyzing them using cultural memory theory. Throughout his narrative of Julio-Claudian Rome in the Annals, Tacitus includes numerous references to the gods, fate, fortune, astrology, omens, temples, priests, emperor cult, and other religious material. Tacitus, who was not only a historian but also a member of Rome’s quindecimviral priesthood, shows a marked interest in even the most detailed rituals of Roman religious life. Yet his portrayal of religious material also suggests that the system is under threat with the advent of the principate. Traditional rituals are forgotten as the shape of the Roman state changes. Simultaneously, a new form of cultic commemoration develops as deceased emperors are deified and the living emperor and his family members are treated in increasingly worshipful ways by his subjects. The study traces the deployment of religious material throughout Tacitus’ narrative, to show how Tacitus views the development of this cultic ‘amnesia’ over time, from the reign of the cryptic, autocratic, and oddly mystical Tiberius, through Claudius’ failed attempts at reviving tradition, to the final sacrilegious disasters of the impious Nero.

Sign in / Sign up

Export Citation Format

Share Document