arXiv subjects

H. V. Jagadish

Publications and source records attributed to H. V. Jagadish.

46 records · Page 3Linked to original sources

arXiv2018

A Generic Inverted Index Framework for Similarity Search on the GPU - Technical Report

We propose a novel generic inverted index framework on the GPU (called GENIE), aiming to reduce the programming complexity of the GPU for parallel similarity search of different data types. Not every data type and similarity measure are supported by GENIE, but many popular ones are. We present the system design of GENIE, and demonstrate similarity search with GENIE on several data types along with a theoretical analysis of search results. A new concept of locality sensitive hashing (LSH) named $τ$-ANN search, and a novel data structure c-PQ on the GPU are also proposed for achieving this purpose. Extensive experiments on different real-life datasets demonstrate the efficiency and effectiveness of our framework. The implemented system has been released as open source.

cs.DB↗

arXiv2018

PANDA: Facilitating Usable AI Development

Recent advances in artificial intelligence (AI) and machine learning have created a general perception that AI could be used to solve complex problems, and in some situations over-hyped as a tool that can be so easily used. Unfortunately, the barrier to realization of mass adoption of AI on various business domains is too high because most domain experts have no background in AI. Developing AI applications involves multiple phases, namely data preparation, application modeling, and product deployment. The effort of AI research has been spent mostly on new AI models (in the model training stage) to improve the performance of benchmark tasks such as image recognition. Many other factors such as usability, efficiency and security of AI have not been well addressed, and therefore form a barrier to democratizing AI. Further, for many real world applications such as healthcare and autonomous driving, learning via huge amounts of possibility exploration is not feasible since humans are involved. In many complex applications such as healthcare, subject matter experts (e.g. Clinicians) are the ones who appreciate the importance of features that affect health, and their knowledge together with existing knowledge bases are critical to the end results. In this paper, we take a new perspective on developing AI solutions, and present a solution for making AI usable. We hope that this resolution will enable all subject matter experts (eg. Clinicians) to exploit AI like data scientists.

cs.AI↗

arXiv2018

RRR: Rank-Regret Representative

Selecting the best items in a dataset is a common task in data exploration. However, the concept of "best" lies in the eyes of the beholder: different users may consider different attributes more important, and hence arrive at different rankings. Nevertheless, one can remove "dominated" items and create a "representative" subset of the data set, comprising the "best items" in it. A Pareto-optimal representative is guaranteed to contain the best item of each possible ranking, but it can be almost as big as the full data. Representative can be found if we relax the requirement to include the best item for every possible user, and instead just limit the users' "regret". Existing work defines regret as the loss in score by limiting consideration to the representative instead of the full data set, for any chosen ranking function. However, the score is often not a meaningful number and users may not understand its absolute value. Sometimes small ranges in score can include large fractions of the data set. In contrast, users do understand the notion of rank ordering. Therefore, alternatively, we consider the position of the items in the ranked list for defining the regret and propose the {\em rank-regret representative} as the minimal subset of the data containing at least one of the top-$k$ of any possible ranking function. This problem is NP-complete. We use the geometric interpretation of items to bound their ranks on ranges of functions and to utilize combinatorial geometry notions for developing effective and efficient approximation algorithms for the problem. Experiments on real datasets demonstrate that we can efficiently find small subsets with small rank-regrets.

cs.DB↗

arXiv2018

Designing Fair Ranking Schemes

Items from a database are often ranked based on a combination of multiple criteria. A user may have the flexibility to accept combinations that weigh these criteria differently, within limits. On the other hand, this choice of weights can greatly affect the fairness of the produced ranking. In this paper, we develop a system that helps users choose criterion weights that lead to greater fairness. We consider ranking functions that compute the score of each item as a weighted sum of (numeric) attribute values, and then sort items on their score. Each ranking function can be expressed as a vector of weights, or as a point in a multi-dimensional space. For a broad range of fairness criteria, we show how to efficiently identify regions in this space that satisfy these criteria. Using this identification method, our system is able to tell users whether their proposed ranking function satisfies the desired fairness criteria and, if it does not, to suggest the smallest modification that does. We develop user-controllable approximation that and indexing techniques that are applied during preprocessing, and support sub-second response times during the online phase. Our extensive experiments on real datasets demonstrate that our methods are able to find solutions that satisfy fairness criteria effectively and efficiently.

cs.DB↗

arXiv2017

Bsmooth: Learning from user feedback to disambiguate query terms in interactive data retrieval

There is great interest in supporting imprecise queries (e.g., keyword search or natural language queries) over databases today. To support such queries, the database system is typically required to disambiguate parts of the user-specified query against the database, using whatever resources are intrinsically available to it (the database schema, data values distributions, natural language models etc). Often, systems will also have a user-interaction log available, which can serve as an extrinsic resource to supplement their model based on their own intrinsic resources. This leads to a problem of how best to combine the system's prior ranking with insight derived from the user-interaction log. Statistical inference techniques such as maximum likelihood or Bayesian updates from a subjective prior turn out not to apply in a straightforward way due to possible noise from user search behavior and to encoding biases endemic to the system's models. In this paper, we address such learning problem in interactive data retrieval, with specific focus on type classification for user-specified query terms. We develop a novel Bayesian smoothing algorithm, Bsmooth, which is simple, fast, flexible and accurate. We analytically establish some desirable properties and show, through experiments against an independent benchmark, that the addition of such a learning layer performs much better than standard methods.

cs.DB↗

arXiv2016

Tuning Crowdsourced Human Computation

As the use of crowdsourcing increases, it is important to think about performance optimization. For this purpose, it is possible to think about each worker as a HPU(Human Processing Unit), and to draw inspiration from performance optimization on traditional computers or cloud nodes with CPUs. However, as we characterize HPUs in detail for this purpose, we find that there are important differences between CPUs and HPUs, leading to the need for completely new optimization algorithms. In this paper, we study the specific optimization problem of obtaining results fastest for a crowd sourced job with a fixed total budget. In crowdsourcing, jobs are usually broken down into sets of small tasks, which are assigned to workers one at a time. We consider three scenarios of increasing complexity: Identical Round Homogeneous tasks, Multiplex Round Homogeneous tasks, and Multiple Round Heterogeneous tasks. For each scenario, we analyze the stochastic behavior of the HPU clock-rate as a function of the remuneration offered. After that, we develop an optimum Budget Allocation strategy to minimize the latency for job completion. We validate our results through extensive simulations and experiments on Amazon Mechanical Turk.

cs.HC↗

arXiv2016

Cohort Query Processing

Modern Internet applications often produce a large volume of user activity records. Data analysts are interested in cohort analysis, or finding unusual user behavioral trends, in these large tables of activity records. In a traditional database system, cohort analysis queries are both painful to specify and expensive to evaluate. We propose to extend database systems to support cohort analysis. We do so by extending SQL with three new operators. We devise three different evaluation schemes for cohort query processing. Two of them adopt a non-intrusive approach. The third approach employs a columnar based evaluation scheme with optimizations specifically designed for cohort query processing. Our experimental results confirm the performance benefits of our proposed columnar database system, compared against the two non-intrusive approaches that implement cohort queries on top of regular relational databases.

cs.DB↗

arXiv2016

PI : a Parallel in-memory skip list based Index

Due to the coarse granularity of data accesses and the heavy use of latches, indices in the B-tree family are not efficient for in-memory databases, especially in the context of today's multi-core architecture. In this paper, we present PI, a Parallel in-memory skip list based Index that lends itself naturally to the parallel and concurrent environment, particularly with non-uniform memory access. In PI, incoming queries are collected, and disjointly distributed among multiple threads for processing to avoid the use of latches. For each query, PI traverses the index in a Breadth-First-Search (BFS) manner to find the list node with the matching key, exploiting SIMD processing to speed up the search process. In order for query processing to be latch-free, PI employs a light-weight communication protocol that enables threads to re-distribute the query workload among themselves such that each list node that will be modified as a result of query processing will be accessed by exactly one thread. We conducted extensive experiments, and the results show that PI can be up to three times as fast as the Masstree, a state-of-the-art B-tree based index.

cs.DB↗

arXiv2015

Active Sampler: Light-weight Accelerator for Complex Data Analytics at Scale

Recent years have witnessed amazing outcomes from "Big Models" trained by "Big Data". Most popular algorithms for model training are iterative. Due to the surging volumes of data, we can usually afford to process only a fraction of the training data in each iteration. Typically, the data are either uniformly sampled or sequentially accessed. In this paper, we study how the data access pattern can affect model training. We propose an Active Sampler algorithm, where training data with more "learning value" to the model are sampled more frequently. The goal is to focus training effort on valuable instances near the classification boundaries, rather than evident cases, noisy data or outliers. We show the correctness and optimality of Active Sampler in theory, and then develop a light-weight vectorized implementation. Active Sampler is orthogonal to most approaches optimizing the efficiency of large-scale data analytics, and can be applied to most analytics models trained by stochastic gradient descent (SGD) algorithm. Extensive experimental evaluations demonstrate that Active Sampler can speed up the training procedure of SVM, feature selection and deep learning, for comparable training quality by 1.6-2.2x.

cs.DB↗

arXiv1998

The Asilomar Report on Database Research

The database research community is rightly proud of success in basic research, and its remarkable record of technology transfer. Now the field needs to radically broaden its research focus to attack the issues of capturing, storing, analyzing, and presenting the vast array of online data. The database research community should embrace a broader research agenda -- broadening the definition of database management to embrace all the content of the Web and other online data stores, and rethinking our fundamental assumptions in light of technology shifts. To accelerate this transition, we recommend changing the way research results are evaluated and presented. In particular, we advocate encouraging more speculative and long-range work, moving conferences to a poster format, and publishing all research literature on the Web.

cs.DB↗