Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

This paper presents and characterizes the Princeton Application Repository for Shared-Memory Computers (PARSEC), a benchmark suite for studies of Chip-Multiprocessors (CMPs). Previous available benchmarks for multiprocessors have focused on high-performance computing applications and used a limited number of synchronization methods. PARSEC includes emerging applications in recognition, mining and synthesis (RMS) as well as systems applications which mimic large-scale multithreaded commercial programs. Our characterization shows that the benchmark suite covers a wide spectrum of working sets, locality, data sharing, synchronization and off-chip traffic. The benchmark suite has been made available to the public.

/pdf/the-parsec-benchmark-suite-characterization-and-jiajft2rem.pdf

The PARSEC benchmark suite: characterization and architectural implications

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI

In recent years, many accurate decision support systems have been constructed as black boxes, that is as systems that hide their internal logic to the user. This lack of explanation constitutes both a practical and an ethical issue. The literature reports many approaches aimed at overcoming this crucial weakness, sometimes at the cost of sacrificing accuracy for interpretability. The applications in which black box decision systems can be used are various, and each approach is typically developed to provide a solution for a specific problem and, as a consequence, it explicitly or implicitly delineates its own definition of interpretability and explanation. The aim of this article is to provide a classification of the main problems addressed in the literature with respect to the notion of explanation and the type of black box system. Given a problem definition, a black box type, and a desired explanation, this survey should help the researcher to find the proposals more useful for his own work. The proposed classification of approaches to open black box models should also be useful for putting the many research open questions in perspective.

https://dl.acm.org/doi/pdf/10.1145/3236009

A Survey of Methods for Explaining Black Box Models

The Social Psychology of Groups. J. W Thibaut & H. H. Kelley. New York: alley, 1959. The team of Thibaut and Kelley goes back to 1946 when, after serving in different units of the armed services psychology program, the authors joined the Research Center for Group Dynamics, first at M.LT and then at the University of Michigan. Their continued association eventuated in appointments as fellows at the Center for Advanced Study in the Behavioral Sciences, 19561957. It is during these years that their collaboration resulted in the publication of The Social Psychology of Groups. The book was designed to "bring order and coherence to present-day research in interpersonal relations and group functioning." To accomplish this aim, the authors introduced, defined, and illustrated basic concepts in an effort to explain the simplest of social phenomena, the two-person relationship. These basic principles and concepts were then employed to illuminate larger problems and more complex social relationships and to examine the significance of such concepts as roles, norm, power, group cohesiveness, and status. The lasting legacy of this book is derived from the fact that the concepts and principles discussed therein serve as a foundation for one of the dominant conceptual frameworks in the field of family studies today-the social exchange framework. Specifically, much of our contemporary thinking about the process of interpersonal attraction and about how individuals evaluate their close relationships has been influenced by the theory and concepts introduced in The Social Psychology of Groups. Today, as a result of Thibaut and Kelley, we think of interpersonal attraction as resulting from the unique valence of driving and restraining forces, rewards and costs, subjectively thought to be available from a specific relationship and its competing alternatives. We understand, as well, that relationships are evaluated through complex and subjectively based comparative processes. As a result, when we think about assessing the degree to which individuals are satisfied with their relationships, we take into consideration the fact that individuals differ in terms of the importance they attribute to different aspects of a relationship (e.g., financial security, sexual fulfillment, companionship). We also take into consideration the fact that individuals differ in terms of the levels of rewards and costs that they believe are realistically obtainable and deserved from a relationship. In addition, as a result of Thibaut and Kelley's theoretical focus on the concept of dependence and the interrelationship between attraction and dependence, there has evolved within the field of family studies a deeper appreciation for the complexities and variability found within relationships. Individuals are dependent on their relationships, according to Thibaut and Kelley, when the outcomes derived from the existing relationship exceed those perceived to be available in competing alternatives. Individuals who are highly dependent on their relationships are less likely to act to end their relationships. This dependence and the stability it engenders may or may not be voluntary, depending on the degree to which individuals are attracted to and satisfied with their relationships. When individuals are both attracted to and dependent on their relationships, they can be thought of as voluntarily participating in their relationship. That is, they are likely to commit themselves to the partner and relationship and actively work for its continuance. Thibaut and Kelley termed those relationships characterized by low levels of satisfaction and high levels of dependence "nonvoluntary relationships. …

The Social Psychology of Groups

Web spam can significantly deteriorate the quality of search engine results. Thus there is a large incentive for commercial search engines to detect spam pages efficiently and accurately. In this paper we present a spam detection system that combines link-based and content-based features, and uses the topology of the Web graph by exploiting the link dependencies among the Web pages. We find that linked hosts tend to belong to the same class: either both are spam or both are non-spam. We demonstrate three methods of incorporating the Web graph topology into the predictions obtained by our base classifier: (i) clustering the host graph, and assigning the label of all hosts in the cluster by majority vote, (ii) propagating the predicted labels to neighboring hosts, and (iii) using the predicted labels of neighboring hosts as new features and retraining the classifier. The result is an accurate system for detecting Web spam, tested on a large and public dataset, using algorithms that can be applied in practice to large-scale Web data.

https://www.cc.gatech.edu/~lingliu/courses/cs4440/07Fall/p423-castillo.pdf

Know your neighbors: web spam detection using the web topology

In this paper we study the trade-offs in designing efficient caching systems for Web search engines. We explore the impact of different approaches, such as static vs. dynamic caching, and caching query results vs.caching posting lists. Using a query log spanning a whole year we explore the limitations of caching and we demonstrate that caching posting lists can achieve higher hit rates than caching query answers. We propose a new algorithm for static caching of posting lists, which outperforms previous methods. We also study the problem of finding the optimal way to split the static cache between answers and posting lists. Finally, we measure how the changes in the query log affect the effectiveness of static caching, given our observation that the distribution of the queries changes slowly over time. Our results and observations are applicable to different levels of the data-access hierarchy, for instance, for a memory/disk layer or a broker/remote server layer.

The impact of caching on search engines

This article discusses efficiency and effectiveness issues in caching the results of queries submitted to a Web search engine (WSE). We propose SDC (Static Dynamic Cache), a new caching strategy aimed to efficiently exploit the temporal and spatial locality present in the stream of processed queries. SDC extracts from historical usage data the results of the most frequently submitted queries and stores them in a static, read-only portion of the cache. The remaining entries of the cache are dynamically managed according to a given replacement policy and are used for those queries that cannot be satisfied by the static portion. Moreover, we improve the hit ratio of SDC by using an adaptive prefetching strategy, which anticipates future requests by introducing a limited overhead over the back-end WSE. We experimentally demonstrate the superiority of SDC over purely static and dynamic policies by measuring the hit ratio achieved on three large query logs by varying the cache parameters and the replacement policy used for managing the dynamic part of the cache. Finally, we deploy and measure the throughput achieved by a concurrent version of our caching system. Our tests show how the SDC cache can be efficiently exploited by many threads that concurrently serve the queries of different users.

/pdf/boosting-the-performance-of-web-search-engines-caching-and-4x5ek32k0w.pdf

Boosting the performance of Web search engines: Caching and prefetching query results by exploiting historical usage data

Web search engines have stored in their logs information about users since they started to operate. This information often serves many purposes. The primary focus of this survey is on introducing to the discipline of query mining by showing its foundations and by analyzing the basic algorithms and techniques that are used to extract useful knowledge from this (potentially) infinite source of information. We show how search applications may benefit from this kind of analysis by analyzing popular applications of query log mining and their influence on user experience. We conclude the paper by, briefly, presenting some of the most challenging current open problems in this field.

/pdf/mining-query-logs-turning-search-usage-data-into-knowledge-2qnudn6t8k.pdf

Mining Query Logs: Turning Search Usage Data into Knowledge

Given the large number of installed apps and the limited screen size of mobile devices, it is often tedious for users to search for the app they want to use. Although some mobile OSs provide categorization schemes that enhance the visibility of useful apps among those installed, the emerging category of homescreen apps aims to take one step further by automatically organizing the installed apps in a more intelligent and personalized way. In this paper, we study how to improve homescreen apps' usage experience through a prediction mechanism that allows to show to users which app she is going to use in the immediate future. The prediction technique is based on a set of features representing the real-time spatiotemporal contexts sensed by the homescreen app. We model the prediction of the next app as a classification problem and propose an effective personalized method to solve it that takes full advantage of human-engineered features and automatically derived features. Furthermore, we study how to solve the two naturally associated cold-start problems: app cold-start and user cold-start. We conduct large-scale experiments on log data obtained from Yahoo Aviate, showing that our approach can accurately predict the next app that a person is going to use.

/pdf/predicting-the-next-app-that-you-are-going-to-use-2xzcqxov6z.pdf

Fabrizio Silvestri

Papers

Know your neighbors: web spam detection using the web topology

The impact of caching on search engines

Boosting the performance of Web search engines: Caching and prefetching query results by exploiting historical usage data

Mining Query Logs: Turning Search Usage Data into Knowledge

Predicting The Next App That You Are Going To Use