An official website of the United States government
The problem of handling adaptivity in data analysis, intentional or not, permeates a variety of fields, including test-set overfitting in ML challenges and the accumulation of invalid scientific discoveries. We propose a mechanism for answering an arbitrarily long sequence of potentially adaptive statistical queries, by charging a price for each query and using the proceeds to collect additional samples. Crucially, we guarantee statistical validity without any assumptions on how the queries are generated. We also ensure with high probability that the cost for M non-adaptive queries is O(log M), while the cost to a potentially adaptive user who makes M queries that do not depend on any others is O(sqrt(M)).
more »
« less
Program Analysis for Adaptive Data Analysis
Data analyses are usually designed to identify some property of the population from which the data are drawn, generalizing beyond the specific data sample. For this reason, data analyses are often designed in a way that guarantees that they produce a low generalization error. That is, they are designed so that the result of a data analysis run on a sample data does not differ too much from the result one would achieve by running the analysis over the entire population. An adaptive data analysis can be seen as a process composed by multiple queries interrogating some data, where the choice of which query to run next may rely on the results of previous queries. The generalization error of each individual query/analysis can be controlled by using an array of well-established statistical techniques. However, when queries are arbitrarily composed, the different errors can propagate through the chain of different queries and bring to a high generalization error. To address this issue, data analysts are designing several techniques that not only guarantee bounds on the generalization errors of single queries, but that also guarantee bounds on the generalization error of the composed analyses. The choice of which of these techniques to use, often depends on the chain of queries that an adaptive data analysis can generate. In this work, we consider adaptive data analyses implemented as while-like programs and we design a program analysis which can help with identifying which technique to use to control their generalization errors. More specifically, we formalize the intuitive notion ofadaptivityas a quantitative property of programs. We do this because the adaptivity level of a data analysis is a key measure to choose the right technique. Based on this definition, we design a program analysis for soundly approximating this quantity. The program analysis generates a representation of the data analysis as a weighted dependency graph, where the weight is an upper bound on the number of times each variable can be reached, and uses a path search strategy to guarantee an upper bound on the adaptivity. We implement our program analysis and show that it can help to analyze the adaptivity of several concrete data analyses with different adaptivity structures.
more »
« less
- Award ID(s):
- 2040249
- PAR ID:
- 10630404
- Publisher / Repository:
- ACM
- Date Published:
- Journal Name:
- Proceedings of the ACM on Programming Languages
- Volume:
- 8
- Issue:
- PLDI
- ISSN:
- 2475-1421
- Page Range / eLocation ID:
- 914 to 938
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Filters trade off accuracy for space and occasionally return false positive matches with a bounded error. Numerous systems use filters in fast memory to avoid performing expensive I/Os to slow storage. A fundamental limitation in traditional filters is that they do not change their representation upon seeing a false positive match. Therefore, the maximum false positive rate is only guaranteed for a single query, not for an arbitrary set of queries. We can improve the filter's performance on a stream of queries, especially on a skewed distribution, if we can adapt after encountering false positives. Adaptive filters, such as telescoping quotient filters and adaptive cuckoo filters, update their representation upon detecting a false positive to avoid repeating the same error in the future. Adaptive filters require an auxiliary structure, typically much larger than the main filter and often residing on slow storage, to facilitate adaptation. However, existing adaptive filters are not practical and have not been adopted in real-world systems for two main reasons. First, they offer weak adaptivity guarantees, meaning that fixing a new false positive can cause a previously fixed false positive to come back. Secondly, the sub-optimal design of the auxiliary structure results in adaptivity overheads so substantial that they can actually diminish overall system performance compared to a traditional filter. In this paper, we design and implement the \sysname, the first practical adaptive filter with minimal adaptivity overhead and strong adaptivity guarantees, which means that the performance and false-positive guarantees continue to hold even for adversarial workloads. The \sysname is based on the state-of-the-art quotient filter design and preserves all the critical features of the quotient filter such as cache efficiency and mergeability. Furthermore, we employ a new auxiliary structure design which results in considerably low adaptivity overhead and makes the \sysname practical in real systems. We evaluate the \sysname by using it to filter queries to an on-disk B-tree database and find no negative impact on insert or query performance compared to traditional filters. Against adversarial workloads, the \sysname preserves system performance, whereas traditional filters incur 2× slowdown from adversaries representing as low as 1% of the workload. Finally, we show that on skewed query workloads, the \sysname can reduce the false-positive rate 100× using negligible (1/1000th of a bit per item) space overhead.more » « less
-
Backward error analysisoffers a method for assessing the quality of numerical programs in the presence of floating-point rounding errors. However, techniques from the numerical analysis literature for quantifying backward error require substantial human effort, and there are currently no tools or automated methods for statically deriving sound backward error bounds. To address this gap, we propose Bean, a typed first-order programming language designed to express quantitative bounds on backward error. Bean’s type system combines a graded coeffect system with strict linearity to soundly track the flow of backward error through programs. We prove the soundness of our system using a novel categorical semantics, where every Bean program denotes a triple of related transformations that together satisfy a backward error guarantee. To illustrate Bean’s potential as a practical tool for automated backward error analysis, we implement a variety of standard algorithms from numerical linear algebra in Bean, establishing fine-grained backward error bounds via typing in a compositional style. We also develop a prototype implementation of Bean that infers backward error bounds automatically. Our evaluation shows that these inferred bounds match worst-case theoretical relative backward error bounds from the literature, underscoring Bean’s utility in validating a key property of numerical programs:numerical stability.more » « less
-
While standard statistical inference techniques and machine learning generalization bounds assume that tests are run on data selected independently of the hypotheses, practical data analysis and machine learning are usually iterative and adaptive processes where the same holdout data is often used for testing a sequence of hypotheses (or models), which may each depend on the outcome of the previous tests on the same data. In this work, we present RADABOUND a rigorous, efficient and practical procedure for controlling the generalization error when using a holdout sample for multiple adaptive testing. Our solution is based on a new application of the Rademacher Complexity generalization bounds, adapted to dependent tests. We demonstrate the statistical power and practicality of our method through extensive simulations and comparisons to alternative approaches. In particular, we show that our rigorous solution is a substantially more powerful and efficient than the differential privacy based approach proposed in Dwork et al. [1]-[3].more » « less
-
We investigate model assessment and selection in a changing environment, by synthesizing datasets from both the current time period and historical epochs. To tackle unknown and potentially arbitrary temporal distribution shift, we develop an adaptive rolling window approach to estimate the generalization error of a given model. This strategy also facilitates the comparison between any two candidate models by estimating the difference of their generalization errors. We further integrate pairwise comparisons into a single-elimination tournament, achieving near-optimal model selection from a collection of candidates. Theoretical analyses and empirical experiments underscore the adaptivity of our proposed methods to the nonstationarity in data.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1145/3656414
