arXiv subjects

Henry Pfister

Publications and source records attributed to Henry Pfister.

6 recordsLinked to original sources

arXiv2023

Attacks on Perception-Based Control Systems: Modeling and Fundamental Limits

We study the performance of perception-based control systems in the presence of attacks, and provide methods for modeling and analysis of their resiliency to stealthy attacks on both physical and perception-based sensing. Specifically, we consider a general setup with a nonlinear affine physical plant controlled with a perception-based controller that maps both the physical (e.g., IMUs) and perceptual (e.g., camera) sensing to the control input; the system is also equipped with a statistical or learning-based anomaly detector (AD). We model the attacks in the most general form, and introduce the notions of attack effectiveness and stealthiness independent of the used AD. In such setting, we consider attacks with different levels of runtime knowledge about the plant. We find sufficient conditions for existence of stealthy effective attacks that force the plant into an unsafe region without being detected by any AD. We show that as the open-loop unstable plant dynamics diverges faster and the closed-loop system converges faster to an equilibrium point, the system is more vulnerable to effective stealthy attacks. Also, depending on runtime information available to the attacker, the probability of attack remaining stealthy can be arbitrarily close to one, if the attacker's estimate of the plant's state is arbitrarily close to the true state; when an accurate estimate of the plant state is not available, the stealthiness level depends on the control performance in attack-free operation.

eess.SY↗

arXiv2021

Hölder Bounds for Sensitivity Analysis in Causal Reasoning

We examine interval estimation of the effect of a treatment T on an outcome Y given the existence of an unobserved confounder U. Using Hölder's inequality, we derive a set of bounds on the confounding bias |E[Y|T=t]-E[Y|do(T=t)]| based on the degree of unmeasured confounding (i.e., the strength of the connection U->T, and the strength of U->Y). These bounds are tight either when U is independent of T or when U is independent of Y given T (when there is no unobserved confounding). We focus on a special case of this bound depending on the total variation distance between the distributions p(U) and p(U|T=t), as well as the maximum (over all possible values of U) deviation of the conditional expected outcome E[Y|U=u,T=t] from the average expected outcome E[Y|T=t]. We discuss possible calibration strategies for this bound to get interval estimates for treatment effects, and experimentally validate the bound using synthetic and semi-synthetic datasets.

cs.LG↗

arXiv2019

Understanding Phase Transitions via Mutual Information and MMSE

The ability to understand and solve high-dimensional inference problems is essential for modern data science. This article examines high-dimensional inference problems through the lens of information theory and focuses on the standard linear model as a canonical example that is both rich enough to be practically useful and simple enough to be studied rigorously. In particular, this model can exhibit phase transitions where an arbitrarily small change in the model parameters can induce large changes in the quality of estimates. For this model, the performance of optimal inference can be studied using the replica method from statistical physics but, until recently, it was not known if the resulting formulas were actually correct. In this chapter, we present a tutorial description of the standard linear model and its connection to information theory. We also describe the replica prediction for this model and outline the authors' recent proof that it is exact.

cs.IT↗

arXiv2018

Adversarial Defense of Image Classification Using a Variational Auto-Encoder

Deep neural networks are known to be vulnerable to adversarial attacks. This exposes them to potential exploits in security-sensitive applications and highlights their lack of robustness. This paper uses a variational auto-encoder (VAE) to defend against adversarial attacks for image classification tasks. This VAE defense has a few nice properties: (1) it is quite flexible and its use of randomness makes it harder to attack; (2) it can learn disentangled representations that prevent blurry reconstruction; and (3) a patch-wise VAE defense strategy is used that does not require retraining for different size images. For moderate to severe attacks, this system outperforms or closely matches the performance of JPEG compression, with the best quality parameter. It also has more flexibility and potential for improvement via training.

cs.CV↗

arXiv2012

On The Performance of Random Block Codes over Finite-State Fading Channels

As the mobile application landscape expands, wireless networks are tasked with supporting various connection profiles, including real-time communications and delay-sensitive traffic. Among many ensuing engineering challenges is the need to better understand the fundamental limits of forward error correction in non-asymptotic regimes. This article seeks to characterize the performance of block codes over finite-state channels with memory. In particular, classical results from information theory are revisited in the context of channels with rate transitions, and bounds on the probabilities of decoding failure are derived for random codes. This study offers new insights about the potential impact of channel correlation over time on overall performance.

cs.IT↗

arXiv2008

Joint Physical Layer Coding and Network Coding for Bi-Directional Relaying

We consider the problem of two transmitters wishing to exchange information through a relay in the middle. The channels between the transmitters and the relay are assumed to be synchronized, average power constrained additive white Gaussian noise channels with a real input with signal-to-noise ratio (SNR) of snr. An upper bound on the capacity is 1/2 log(1+ snr) bits per transmitter per use of the medium-access phase and broadcast phase of the bi-directional relay channel. We show that using lattice codes and lattice decoding, we can obtain a rate of 1/2 log(0.5 + snr) bits per transmitter, which is essentially optimal at high SNRs. The main idea is to decode the sum of the codewords modulo a lattice at the relay followed by a broadcast phase which performs Slepian-Wolf coding with structured codes. For asymptotically low SNR's, jointly decoding the two transmissions at the relay (MAC channel) is shown to be optimal. We also show that if the two transmitters use identical lattices with minimum angle decoding, we can achieve the same rate of 1/2 log(0.5 + snr). The proposed scheme can be thought of as a joint physical layer, network layer code which outperforms other recently proposed analog network coding schemes.

cs.IT↗