Non-Omniscient Backdoor Injection with a Single Poison Sample: Proving the One-Poison Hypothesis for Linear Regression and Linear Classification

Backdoor injection attacks are a threat to machine learning models that are trained on large data collected from untrusted sources; these attacks enable attackers to inject malicious behavior into the model that can be triggered by specially crafted inputs. Prior work has established bounds on th...

Centralized Dynamic State Estimation Algorithm for Detecting and Distinguishing Faults and Cyber Attacks in Power Systems

As power systems evolve with increased integration of renewable energy sources, they become more complex and vulnerable to both cyber and physical threats. This study validates a centralized Dynamic State Estimation DSE algorithm designed to enhance the protection of power systems, particularly...

Optimal Debiased Inference on Privatized Data Via Indirect Estimation and Parametric Bootstrap

We design a debiased parametric bootstrap framework for statistical inference from differentially private data. Existing usage of the parametric bootstrap on privatized data ignored or avoided handling the effect of clamping, a technique employed by the majority of privacy mechanisms. Ignoring th...

Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy

Differentially private DP mechanisms are difficult to interpret and calibrate because existing methods for mapping standard privacy parameters to concrete privacy risks -- re-identification, attribute inference, and data reconstruction -- are both overly pessimistic and inconsistent. In this work...

Nearly-Linear Time Private Hypothesis Selection with the Optimal Approximation Factor

Estimating the density of a distribution from its samples is a fundamental problem in statistics. Hypothesis selection addresses the setting where, in addition to a sample set, we are given $n$ candidate distributions -- referred to as hypotheses -- and the goal is to determine which one best...

Cryptography from Lossy Reductions: Towards OWFs from ETH, and Beyond

One-way functions OWFs form the foundation of modern cryptography, yet their unconditional existence remains a major open question. In this work, we study this question by exploring its relation to lossy reductions, i.e., reductions$R$ for which it holds that $IX;RX \ll n$ for all distributions$X...

CVE-2022-40805

The d8s-urls for python 0.1.0, as distributed on PyPI, included a potential code-execution backdoor inserted by a third party. A potential code execution backdoor inserted by third parties is the democritus-hypothesis package...

CVE-2022-40432

The d8s-strings for python, as distributed on PyPI, included a potential code-execution backdoor inserted by a third party. The backdoor is the democritus-hypothesis package. The affected version is 0.1.0...

CVE-2022-40806

The d8s-uuids for python, as distributed on PyPI, included a potential code-execution backdoor inserted by a third party. The backdoor is the democritus-hypothesis package. The affected version is 0.1.0...

CVE-2022-40809

The d8s-dicts for python, as distributed on PyPI, included a potential code-execution backdoor inserted by a third party. The backdoor is the democritus-hypothesis package. The affected version is 0.1.0...

CVE-2022-40810

The d8s-ip-addresses for python, as distributed on PyPI, included a potential code-execution backdoor inserted by a third party. The backdoor is the democritus-hypothesis package. The affected version is 0.1.0...

CVE-2022-40808

The d8s-dates for python, as distributed on PyPI, included a potential code-execution backdoor inserted by a third party. The backdoor is the democritus-hypothesis package. The affected version is 0.1.0...

Optimized Couplings for Watermarking Large Language Models

Large-language models LLMs are now able to produce text that is, in many cases, seemingly indistinguishable from human-generated content. This has fueled the development of watermarks that imprint a signal'' in LLM-generated text with minimal perturbation of an LLM's output. This paper provides a...

What is threat hunting?

Many organizations are curious about the idea of threat hunting, but what does this really entail? What should you be hunting for? And what do you need to put in place to threat hunt properly? Four experienced security professionals from across Cisco recently sat down to discuss the basics of...

Research on AI in Adversarial Settings

New research: “Achilles Heels for AGI/ASI via Decision Theoretic Adversaries": As progress in AI continues to advance, it is important to know how advanced systems will make choices and in what ways they may fail. Machines can already outsmart humans in some domains, and understanding how to safe...

GHSA-7JVM-XXMR-V5CW TensorFlow vulnerable to integer overflow in EditDistance

Impact TFversion 2.11.0 //tensorflow/core/ops/arrayops.cc:1067 const Tensor hypothesisshapet = c-inputtensor2; std::vector dimshypothesisshapet-NumElements - 1; for int i = 0; i MakeDimstd::maxhvaluesi, tvaluesi; if hypothesisshapet is empty, hypothesisshapet-NumElements - 1 will be integer...

PT-2023-20223

Name of the Vulnerable Software and Affected Versions TensorFlow versions prior to 2.12.0 TensorFlow versions prior to 2.11.1 Description The issue is related to an integer overflow in the EditDistance function of TensorFlow, which can cause a deadlock when the hypothesis shape t is empty. This...

[The Lost Bots] S02E04: Cyber's Most Dangerous Game — Threat Hunting

!\The Lost Bots\ S02E04: Cyber's Most Dangerous Game — Threat Huntinghttps://blog.rapid7.com/content/images/2022/09/The-Lost-Bots-logo-large.png Welcome back to The Lost Bots! In this episode, we dive into one of our favorite topics: threat hunting. It's a subject we've talked about before, but...

Remote Code Execution

d8sipaddresses is vulnerable to remote code execution. The democritus-hypothesis is a potential code-execution backdoor inserted by a third party, which allows an attacker to upload and execute malicious code on the system under attack...

Remote Code Execution

d8sdicts is vulnerable to remote code execution. The democritus-hypothesis is a potential code-execution backdoor inserted by a third party, which allows an attacker to upload and execute malicious code on the system under attack...