arXiv subjects

Jirong Yi

Publications and source records attributed to Jirong Yi.

18 recordsLinked to original sources

arXiv2025

Feature compression is the root cause of adversarial fragility in neural network classifiers

In this paper, we uniquely study the adversarial robustness of deep neural networks (NN) for classification tasks against that of optimal classifiers. We look at the smallest magnitude of possible additive perturbations that can change a classifier's output. We provide a matrix-theoretic explanation of the adversarial fragility of deep neural networks for classification. In particular, our theoretical results show that a neural network's adversarial robustness can degrade as the input dimension $d$ increases. Analytically, we show that neural networks' adversarial robustness can be only $1/\sqrt{d}$ of the best possible adversarial robustness of optimal classifiers. Our theories match remarkably well with numerical experiments of practically trained NN, including NN for ImageNet images. The matrix-theoretic explanation is consistent with an earlier information-theoretic feature-compression-based explanation for the adversarial fragility of neural networks.

cs.LG↗

arXiv2023

Outlier Detection Using Generative Models with Theoretical Performance Guarantees

This paper considers the problem of recovering signals modeled by generative models from linear measurements contaminated with sparse outliers. We propose an outlier detection approach for reconstructing the ground-truth signals modeled by generative models under sparse outliers. We establish theoretical recovery guarantees for reconstruction of signals using generative models in the presence of outliers, giving lower bounds on the number of correctable outliers. Our results are applicable to both linear generator neural networks and the nonlinear generator neural networks with an arbitrary number of layers. We propose an iterative alternating direction method of multipliers (ADMM) algorithm for solving the outlier detection problem via $\ell_1$ norm minimization, and a gradient descent algorithm for solving the outlier detection problem via squared $\ell_1$ norm minimization. We conduct extensive experiments using variational auto-encoder and deep convolutional generative adversarial networks, and the experimental results show that the signals can be successfully reconstructed under outliers using our approach. Our approach outperforms the traditional Lasso and $\ell_2$ minimization approach.

stat.ML↗

arXiv2022

Mutual Information Learned Regressor: an Information-theoretic Viewpoint of Training Regression Systems

As one of the central tasks in machine learning, regression finds lots of applications in different fields. An existing common practice for solving regression problems is the mean square error (MSE) minimization approach or its regularized variants which require prior knowledge about the models. Recently, Yi et al., proposed a mutual information based supervised learning framework where they introduced a label entropy regularization which does not require any prior knowledge. When applied to classification tasks and solved via a stochastic gradient descent (SGD) optimization algorithm, their approach achieved significant improvement over the commonly used cross entropy loss and its variants. However, they did not provide a theoretical convergence analysis of the SGD algorithm for the proposed formulation. Besides, applying the framework to regression tasks is nontrivial due to the potentially infinite support set of the label. In this paper, we investigate the regression under the mutual information based supervised learning framework. We first argue that the MSE minimization approach is equivalent to a conditional entropy learning problem, and then propose a mutual information learning formulation for solving regression problems by using a reparameterization technique. For the proposed formulation, we give the convergence analysis of the SGD algorithm for solving it in practice. Finally, we consider a multi-output regression data model where we derive the generalization performance lower bound in terms of the mutual information associated with the underlying data distribution. The result shows that the high dimensionality can be a bless instead of a curse, which is controlled by a threshold. We hope our work will serve as a good starting point for further research on the mutual information based regression.

stat.ML↗

arXiv2022

Mutual Information Learned Classifiers: an Information-theoretic Viewpoint of Training Deep Learning Classification Systems

Deep learning systems have been reported to acheive state-of-the-art performances in many applications, and one of the keys for achieving this is the existence of well trained classifiers on benchmark datasets which can be used as backbone feature extractors in downstream tasks. As a main-stream loss function for training deep neural network (DNN) classifiers, the cross entropy loss can easily lead us to find models which demonstrate severe overfitting behavior when no other techniques are used for alleviating it such as data augmentation. In this paper, we prove that the existing cross entropy loss minimization for training DNN classifiers essentially learns the conditional entropy of the underlying data distribution of the dataset, i.e., the information or uncertainty remained in the labels after revealing the input. In this paper, we propose a mutual information learning framework where we train DNN classifiers via learning the mutual information between the label and input. Theoretically, we give the population error probability lower bound in terms of the mutual information. In addition, we derive the mutual information lower and upper bounds for a concrete binary classification data model in $\mbR^n$, and also the error probability lower bound in this scenario. Besides, we establish the sample complexity for accurately learning the mutual information from empirical data samples drawn from the underlying data distribution. Empirically, we conduct extensive experiments on several benchmark datasets to support our theory. Without whistles and bells, the proposed mutual information learned classifiers (MILCs) acheive far better generalization performances than the state-of-the-art classifiers with an improvement which can exceed more than 10\% in testing accuracy.

cs.LG↗

arXiv2022

Mutual Information Learned Classifiers: an Information-theoretic Viewpoint of Training Deep Learning Classification Systems

Deep learning systems have been reported to achieve state-of-the-art performances in many applications, and a key is the existence of well trained classifiers on benchmark datasets. As a main-stream loss function, the cross entropy can easily lead us to find models which demonstrate severe overfitting behavior. In this paper, we show that the existing cross entropy loss minimization problem essentially learns the label conditional entropy (CE) of the underlying data distribution of the dataset. However, the CE learned in this way does not characterize well the information shared by the label and the input. In this paper, we propose a mutual information learning framework where we train deep neural network classifiers via learning the mutual information between the label and the input. Theoretically, we give the population classification error lower bound in terms of the mutual information. In addition, we derive the mutual information lower and upper bounds for a concrete binary classification data model in $\mathbb{R}^n$, and also the error probability lower bound in this scenario. Empirically, we conduct extensive experiments on several benchmark datasets to support our theory. The mutual information learned classifiers (MILCs) achieve far better generalization performances than the conditional entropy learned classifiers (CELCs) with an improvement which can exceed more than 10\% in testing accuracy.

cs.LG↗

arXiv2021

Optimal Needle Placement for Prostate Rotating-Shield Brachytherapy (RSBT)

Purpose: To present an efficient NEEdle Position Optimization (NEEPO) algorithm for prostate rotating shield brachytherapy (RSBT). With RSBT, the increased flexibility beyond conventional high-dose-rate brachytherapy (HDR-BT) due to the partially shielded radiation source has been shown by Adams et al. in 2020 to enable improved urethra sparing (23.1%), enhanced dose escalation (29.9%), or both, with 20 needles without NEEPO-optimized positions. Within this regime of improved dosimetry, we propose in this work that the benefits of RSBT can be maintained while also reducing the number of needles needed for the delivery. The goal of NEEPO is to provide the capability to further increase the dosimetric benefit of RSBT and to minimize the number of needles needed to satisfy a dosimetric goal. Methods: The NEEPO algorithm generates a needle pool for a given patient and then iteratively constructs a subset of needles from the pool based on relative needle importance as determined by total dwell times within needles. The NEEPO algorithm is based on a convex optimization formulation using a quadratic dosimetric penalty function, dwell time regularization by total variation, and a block sparsity regularization term to enable iterative removal of low-importance needles. RSBT treatment plans for 26 patients were generated using single fraction prescriptions with both dose escalation and urethra sparing goals, and compared to baseline HDR-BT treatment plans.

physics.med-ph↗

arXiv2021

Solving Large Scale Quadratic Constrained Basis Pursuit

Inspired by alternating direction method of multipliers and the idea of operator splitting, we propose a efficient algorithm for solving large-scale quadratically constrained basis pursuit. Experimental results show that the proposed algorithm can achieve 50~~100 times speedup when compared with the baseline interior point algorithm implemented in CVX.

math.OC↗

arXiv2020

Optimal Pooling Matrix Design for Group Testing with Dilution (Row Degree) Constraints

In this paper, we consider the problem of designing optimal pooling matrix for group testing (for example, for COVID-19 virus testing) with the constraint that no more than $r>0$ samples can be pooled together, which we call "dilution constraint". This problem translates to designing a matrix with elements being either 0 or 1 that has no more than $r$ '1's in each row and has a certain performance guarantee of identifying anomalous elements. We explicitly give pooling matrix designs that satisfy the dilution constraint and have performance guarantees of identifying anomalous elements, and prove their optimality in saving the largest number of tests, namely showing that the designed matrices have the largest width-to-height ratio among all constraint-satisfying 0-1 matrices.

q-bio.QM↗

arXiv2020

Error Correction Codes for COVID-19 Virus and Antibody Testing: Using Pooled Testing to Increase Test Reliability

We consider a novel method to increase the reliability of COVID-19 virus or antibody tests by using specially designed pooled testings. Instead of testing nasal swab or blood samples from individual persons, we propose to test mixtures of samples from many individuals. The pooled sample testing method proposed in this paper also serves a different purpose: for increasing test reliability and providing accurate diagnoses even if the tests themselves are not very accurate. Our method uses ideas from compressed sensing and error-correction coding to correct for a certain number of errors in the test results. The intuition is that when each individual's sample is part of many pooled sample mixtures, the test results from all of the sample mixtures contain redundant information about each individual's diagnosis, which can be exploited to automatically correct for wrong test results in exactly the same way that error correction codes correct errors introduced in noisy communication channels. While such redundancy can also be achieved by simply testing each individual's sample multiple times, we present simulations and theoretical arguments that show that our method is significantly more efficient in increasing diagnostic accuracy. In contrast to group testing and compressed sensing which aim to reduce the number of required tests, this proposed error correction code idea purposefully uses pooled testing to increase test accuracy, and works not only in the "undersampling" regime, but also in the "oversampling" regime, where the number of tests is bigger than the number of subjects. The results in this paper run against traditional beliefs that, "even though pooled testing increased test capacity, pooled testings were less reliable than testing individuals separately."

q-bio.QM↗

arXiv2020

Derivation of Information-Theoretically Optimal Adversarial Attacks with Applications to Robust Machine Learning

We consider the theoretical problem of designing an optimal adversarial attack on a decision system that maximally degrades the achievable performance of the system as measured by the mutual information between the degraded signal and the label of interest. This problem is motivated by the existence of adversarial examples for machine learning classifiers. By adopting an information theoretic perspective, we seek to identify conditions under which adversarial vulnerability is unavoidable i.e. even optimally designed classifiers will be vulnerable to small adversarial perturbations. We present derivations of the optimal adversarial attacks for discrete and continuous signals of interest, i.e., finding the optimal perturbation distributions to minimize the mutual information between the degraded signal and a signal following a continuous or discrete distribution. In addition, we show that it is much harder to achieve adversarial attacks for minimizing mutual information when multiple redundant copies of the input signal are available. This provides additional support to the recently proposed ``feature compression" hypothesis as an explanation for the adversarial vulnerability of deep learning classifiers. We also report on results from computational experiments to illustrate our theoretical results.

cs.LG↗

arXiv2020

Low-Cost and High-Throughput Testing of COVID-19 Viruses and Antibodies via Compressed Sensing: System Concepts and Computational Experiments

Coronavirus disease 2019 (COVID-19) is an ongoing pandemic infectious disease outbreak that has significantly harmed and threatened the health and lives of millions or even billions of people. COVID-19 has also negatively impacted the social and economic activities of many countries significantly. With no approved vaccine available at this moment, extensive testing of COVID-19 viruses in people are essential for disease diagnosis, virus spread confinement, contact tracing, and determining right conditions for people to return to normal economic activities. Identifying people who have antibodies for COVID-19 can also help select persons who are suitable for undertaking certain essential activities or returning to workforce. However, the throughputs of current testing technologies for COVID-19 viruses and antibodies are often quite limited, which are not sufficient for dealing with COVID-19 viruses' anticipated fast oscillating waves of spread affecting a significant portion of the earth's population. In this paper, we propose to use compressed sensing (group testing can be seen as a special case of compressed sensing when it is applied to COVID-19 detection) to achieve high-throughput rapid testing of COVID-19 viruses and antibodies, which can potentially provide tens or even more folds of speedup compared with current testing technologies. The proposed compressed sensing system for high-throughput testing can utilize expander graph based compressed sensing matrices developed by us \cite{Weiyuexpander2007}.

q-bio.QM↗

arXiv2020

Do Deep Minds Think Alike? Selective Adversarial Attacks for Fine-Grained Manipulation of Multiple Deep Neural Networks

Recent works have demonstrated the existence of {\it adversarial examples} targeting a single machine learning system. In this paper we ask a simple but fundamental question of "selective fooling": given {\it multiple} machine learning systems assigned to solve the same classification problem and taking the same input signal, is it possible to construct a perturbation to the input signal that manipulates the outputs of these {\it multiple} machine learning systems {\it simultaneously} in arbitrary pre-defined ways? For example, is it possible to selectively fool a set of "enemy" machine learning systems but does not fool the other "friend" machine learning systems? The answer to this question depends on the extent to which these different machine learning systems "think alike". We formulate the problem of "selective fooling" as a novel optimization problem, and report on a series of experiments on the MNIST dataset. Our preliminary findings from these experiments show that it is in fact very easy to selectively manipulate multiple MNIST classifiers simultaneously, even when the classifiers are identical in their architectures, training algorithms and training datasets except for random initialization during training. This suggests that two nominally equivalent machine learning systems do not in fact "think alike" at all, and opens the possibility for many novel applications and deeper understandings of the working principles of deep neural networks.

cs.CV↗

arXiv2019

Trust but Verify: An Information-Theoretic Explanation for the Adversarial Fragility of Machine Learning Systems, and a General Defense against Adversarial Attacks

Deep-learning based classification algorithms have been shown to be susceptible to adversarial attacks: minor changes to the input of classifiers can dramatically change their outputs, while being imperceptible to humans. In this paper, we present a simple hypothesis about a feature compression property of artificial intelligence (AI) classifiers and present theoretical arguments to show that this hypothesis successfully accounts for the observed fragility of AI classifiers to small adversarial perturbations. Drawing on ideas from information and coding theory, we propose a general class of defenses for detecting classifier errors caused by abnormally small input perturbations. We further show theoretical guarantees for the performance of this detection method. We present experimental results with (a) a voice recognition system, and (b) a digit recognition system using the MNIST database, to demonstrate the effectiveness of the proposed defense methods. The ideas in this paper are motivated by a simple analogy between AI classifiers and the standard Shannon model of a communication system.

cs.CR↗

arXiv2019

An Information-Theoretic Explanation for the Adversarial Fragility of AI Classifiers

We present a simple hypothesis about a compression property of artificial intelligence (AI) classifiers and present theoretical arguments to show that this hypothesis successfully accounts for the observed fragility of AI classifiers to small adversarial perturbations. We also propose a new method for detecting when small input perturbations cause classifier errors, and show theoretical guarantees for the performance of this detection method. We present experimental results with a voice recognition system to demonstrate this method. The ideas in this paper are motivated by a simple analogy between AI classifiers and the standard Shannon model of a communication system.

cs.IT↗

arXiv2018

Outlier Detection using Generative Models with Theoretical Performance Guarantees

This paper considers the problem of recovering signals from compressed measurements contaminated with sparse outliers, which has arisen in many applications. In this paper, we propose a generative model neural network approach for reconstructing the ground truth signals under sparse outliers. We propose an iterative alternating direction method of multipliers (ADMM) algorithm for solving the outlier detection problem via $\ell_1$ norm minimization, and a gradient descent algorithm for solving the outlier detection problem via squared $\ell_1$ norm minimization. We establish the recovery guarantees for reconstruction of signals using generative models in the presence of outliers, and give an upper bound on the number of outliers allowed for recovery. Our results are applicable to both the linear generator neural network and the nonlinear generator neural network with an arbitrary number of layers. We conduct extensive experiments using variational auto-encoder and deep convolutional generative adversarial networks, and the experimental results show that the signals can be successfully reconstructed under outliers using our approach. Our approach outperforms the traditional Lasso and $\ell_2$ minimization approach.

cs.IT↗

arXiv2018

Necessary and Sufficient Null Space Condition for Nuclear Norm Minimization in Low-Rank Matrix Recovery

Low-rank matrix recovery has found many applications in science and engineering such as machine learning, signal processing, collaborative filtering, system identification, and Euclidean embedding. But the low-rank matrix recovery problem is an NP hard problem and thus challenging. A commonly used heuristic approach is the nuclear norm minimization. In [12,14,15], the authors established the necessary and sufficient null space conditions for nuclear norm minimization to recover every possible low-rank matrix with rank at most r (the strong null space condition). In addition, in [12], Oymak et al. established a null space condition for successful recovery of a given low-rank matrix (the weak null space condition) using nuclear norm minimization, and derived the phase transition for the nuclear norm minimization. In this paper, we show that the weak null space condition in [12] is only a sufficient condition for successful matrix recovery using nuclear norm minimization, and is not a necessary condition as claimed in [12]. In this paper, we further give a weak null space condition for low-rank matrix recovery, which is both necessary and sufficient for the success of nuclear norm minimization. At the core of our derivation are an inequality for characterizing the nuclear norms of block matrices, and the conditions for equality to hold in that inequality.

math.OC↗

arXiv2017

Separation-Free Super-Resolution from Compressed Measurements is Possible: an Orthonormal Atomic Norm Minimization Approach

We consider the problem of recovering the superposition of $R$ distinct complex exponential functions from compressed non-uniform time-domain samples. Total Variation (TV) minimization or atomic norm minimization was proposed in the literature to recover the $R$ frequencies or the missing data. However, it is known that in order for TV minimization and atomic norm minimization to recover the missing data or the frequencies, the underlying $R$ frequencies are required to be well-separated, even when the measurements are noiseless. This paper shows that the Hankel matrix recovery approach can super-resolve the $R$ complex exponentials and their frequencies from compressed non-uniform measurements, regardless of how close their frequencies are to each other. We propose a new concept of orthonormal atomic norm minimization (OANM), and demonstrate that the success of Hankel matrix recovery in separation-free super-resolution comes from the fact that the nuclear norm of a Hankel matrix is an orthonormal atomic norm. More specifically, we show that, in traditional atomic norm minimization, the underlying parameter values $\textbf{must}$ be well separated to achieve successful signal recovery, if the atoms are changing continuously with respect to the continuously-valued parameter. In contrast, for the OANM, it is possible the OANM is successful even though the original atoms can be arbitrarily close. As a byproduct of this research, we provide one matrix-theoretic inequality of nuclear norm, and give its proof from the theory of compressed sensing.

cs.IT↗

arXiv2017

Fast dose optimization for rotating shield brachytherapy

Purpose: To provide a fast computational method, based on the proximal graph solver (POGS) - a convex optimization solver using the alternating direction method of multipliers (ADMM), for calculating an optimal treatment plan in rotating shield brachytherapy (RSBT). RSBT treatment planning has more degrees of freedom than conventional high-dose-rate brachytherapy (HDR-BT) due to the addition of emission direction, and this necessitates a fast optimization technique to enable clinical usage. // Methods: The multi-helix RSBT (H-RSBT) delivery technique was considered with five representative cervical cancer patients. Treatment plans were generated for all patients using the POGS method and the previously considered commercial solver IBM CPLEX. The rectum, bladder, sigmoid, high-risk clinical target volume (HR-CTV), and HR-CTV boundary were the structures considered in our optimization problem, called the asymmetric dose-volume optimization with smoothness control. Dose calculation resolution was 1x1x3 mm^3 for all cases. The H-RSBT applicator has 6 helices, with 33.3 mm of translation along the applicator per helical rotation and 1.7 mm spacing between dwell positions, yielding 17.5 degree emission angle spacing per 5 mm along the applicator.// Results: For each patient, HR-CTV D90, HR-CTV D100, rectum D2cc, sigmoid D2cc, and bladder D2cc matched within 1% for CPLEX and POGS. Also, we obtained similar EQD2 figures between CPLEX and POGS. POGS was around 18 times faster than CPLEX. Over all patients, total optimization times were 32.1-65.4 seconds for CPLEX and 2.1-3.9 seconds for POGS. // Conclusions: POGS substantially reduced treatment plan optimization time around 18 times for RSBT with similar HR-CTV D90, OAR D2cc values, and EQD2 figure relative to CPLEX, which is significant progress toward clinical translation of RSBT. POGS is also applicable to conventional HDR-BT.

physics.med-ph↗