There has been a flurry of recent literature studying streaming algorithms for which the input stream is chosen adaptively by a blackbox adversary who observes the output of the streaming algorithm at each time step. However, these algorithms fail when the adversary has access to the internal state of the algorithm, rather than just the output of the algorithm. We study streaming algorithms in the whitebox adversarial model, where the stream is
chosen adaptively by an adversary who observes the entire internal state of the algorithm at each time step. We show that nontrivial algorithms are still possible. We first give a randomized algorithm for the L1heavy hitters problem that outperforms the optimal deterministic MisraGries algorithm on long streams. If the whitebox adversary is computationally bounded, we use cryptographic techniques to reduce the memory of our L1heavy hitters algorithm even further
and to design a number of additional algorithms for graph, string, and linear algebra problems. The existence of such algorithms is surprising, as the streaming algorithm does not even have a secret key in this model, i.e., its state is entirely known to the adversary. One algorithm we design is for estimating the number of distinct elements in a stream with insertions and deletions achieving a multiplicative approximation and sublinear space; such an algorithm is impossible for deterministic algorithms. We also give a general technique that translates any twoplayer deterministic communication lower bound to a lower bound for randomized algorithms robust to a whitebox adversary. In particular, our results show that for all p ≥ 0, there exists a constant Cp > 1 such that any
Cpapproximation algorithm for Fp moment estimation in insertiononly streams with a whitebox adversary requires Ω(n) space for a universe of size n. Similarly, there is a constant C > 1 such that any Capproximation algorithm in an insertiononly stream for matrix rank requires Ω(n) space with a whitebox adversary. These results do not contradict our upper bounds since they assume the adversary has unbounded computational power. Our algorithmic results based on cryptography thus show a separation between computationally bounded and unbounded adversaries. Finally, we prove a lower bound of Ω(log n) bits for the fundamental problem of deterministic approximate counting in a stream of 0’s and 1’s, which holds even if we know how many total stream updates we have seen so far at each point in the stream. Such a lower bound for approximate counting with additional information was previously unknown, and in our context, it shows a separation between multiplayer deterministic maximum communication and the whitebox space complexity of a streaming algorithm
more »
« less
A Causal Lens for Peeking into Black Box Predictive Models: Predictive Model Interpretation via Causal Attribution
With the increasing adoption of predictive models trained using machine learning across a wide range of highstakes applications, e.g., health care, security, criminal justice, finance, and education, there is a growing need for effective techniques for explaining such models and their predictions. We aim to address this problem in settings where the predictive model is a black box; That is, we can only observe the response of the model to various inputs, but have no knowledge about the internal structure of the predictive model, its parameters, the objective function, and the algorithm used to optimize the model. We reduce the problem of interpreting a black box predictive model to that of estimating the causal effects of each of the model inputs on the model output, from observations of the model inputs and the corresponding outputs. We estimate the causal effects of model inputs on model output using variants of the Rubin Neyman potential outcomes framework for estimating causal effects from observational data. We show how the resulting causal attribution of responsibility for model output to the different model inputs can be used to interpret the predictive model and to explain its predictions. We present results of experiments that demonstrate the effectiveness of our approach to the interpretation of black box predictive models via causal attribution in the case of deep neural network models trained on one synthetic data set (where the input variables that impact the output variable are known by design) and two realworld data sets: Handwritten digit classification, and Parkinson's disease severity prediction. Because our approach does not require knowledge about the predictive model algorithm and is free of assumptions regarding the black box predictive model except that its inputoutput responses be observable, it can be applied, in principle, to any black box predictive model.
more »
« less
 NSFPAR ID:
 10287271
 Date Published:
 Journal Name:
 ArXivorg
 ISSN:
 23318422
 Page Range / eLocation ID:
 https://arxiv.org/abs/2008.00357
 Format(s):
 Medium: X
 Sponsoring Org:
 National Science Foundation
More Like this


Multilabel classification is a challenging structured prediction task in which a set of output class labels are predicted for each input. Realworld datasets often have natural or latent taxonomic relationships between labels, making it desirable for models to employ label representations capable of capturing such taxonomies. Most existing multilabel classification methods do not do so, resulting in label predictions that are inconsistent with the taxonomic constraints, thus failing to accurately represent the fundamentals of problem setting. In this work, we introduce the multilabel box model (MBM), a multilabel classification method that combines the encoding power of neural networks with the inductive bias and probabilistic semantics of box embeddings (Vilnis, et al 2018). Box embeddings can be understood as trainable Venndiagrams based on hyperrectangles. Representing labels by boxes rather than vectors, MBM is able to capture taxonomic relations among labels. Furthermore, since box embeddings allow these relations to be learned by stochastic gradient descent from data, and to be read as calibrated conditional probabilities, our model is endowed with a high degree of interpretability. This interpretability also facilitates the injection of partial information about labellabel relationships into model training, to further improve its consistency. We provide theoretical grounding for our method and show experimentally the model's ability to learn the true latent taxonomic structure from data. Through extensive empirical evaluations on both small and largescale multilabel classification datasets, we show that BBM can significantly improve taxonomic consistency while preserving or surpassing the stateoftheart predictive performance.more » « less

An important achievement in the field of causal inference was a complete characterization of when a causal effect, in a system modeled by a causal graph, can be determined uniquely from purely observational data. The identification algorithms resulting from this work produce exact symbolic expressions for causal effects, in terms of the observational probabilities. More recent work has looked at the numerical properties of these expressions, in particular using the classical notion of the condition number. In its classical interpretation, the condition number quantifies the sensitivity of the output values of the expressions to small numerical perturbations in the input observational probabilities. In the context of causal identification, the condition number has also been shown to be related to the effect of certain kinds of uncertainties in the structure of the causal graphical model. In this paper, we first give an upper bound on the condition number for the interesting case of causal graphical models with small “confounded components”. We then develop a tight characterization of the condition number of any given causal identification problem. Finally, we use our tight characterization to give a specific example where the condition number can be much lower than that obtained via generic bounds on the condition number, and to show that even “equivalent” expressions for causal identification can behave very differently with respect to their numerical stability properties.more » « less

de Campos, C. ; Maathuis, M. H. (Ed.)An important achievement in the field of causal inference was a complete characterization of when a causal effect, in a system modeled by a causal graph, can be determined uniquely from purely observational data. The identification algorithms resulting from this work produce exact symbolic expressions for causal effects, in terms of the observational probabilities. More recent work has looked at the numerical properties of these expressions, in particular using the classical notion of the condition number. In its classical interpretation, the condition number quantifies the sensitivity of the output values of the expressions to small numerical perturbations in the input observational probabilities. In the context of causal identification, the condition number has also been shown to be related to the effect of certain kinds of uncertainties in the structure of the causal graphical model. In this paper, we first give an upper bound on the condition number for the interesting case of causal graphical models with small “confounded components”. We then develop a tight characterization of the condition number of any given causal identification problem. Finally, we use our tight characterization to give a specific example where the condition number can be much lower than that obtained via generic bounds on the condition number, and to show that even “equivalent” expressions for causal identification can behave very differently with respect to their numerical stability properties.more » « less

Many causal and policy effects of interest are defined by linear functionals of highdimensional or nonparametric regression functions. Rootn consistent and asymptotically normal estimation of the object of interest requires debiasing to reduce the effects of regularization and/or model selection on the object of interest. Debiasing is typically achieved by adding a correction term to the plugin estimator of the functional, which leads to properties such as semiparametric efficiency, double robustness, and Neyman orthogonality. We implement an automatic debiasing procedure based on automatically learning the Riesz representation of the linear functional using Neural Nets and Random Forests. Our method only relies on blackbox evaluation oracle access to the linear functional and does not require knowledge of its analytic form. We propose a multitasking Neural Net debiasing method with stochastic gradient descent minimization of a combined Riesz representer and regression loss, while sharing representation layers for the two functions. We also propose a Random Forest method which learns a locally linear representation of the Riesz function. Even though our method applies to arbitrary functionals, we experimentally find that it performs well compared to the state of art neural net based algorithm of Shi et al. (2019) for the case of the average treatment effect functional. We also evaluate our method on the problem of estimating average marginal effects with continuous treatments, using semisynthetic data of gasoline price changes on gasoline demand. Code available at github.com/victor5as/RieszLearning.more » « less