-
Approximate Distributed Monitoring under Partial Synchrony: Balancing Speed and Accuracy
Authors:
Borzoo Bonakdarpour,
Anik Momtaz,
Dejan Ničković,
N. Ege Saraç
Abstract:
In distributed systems with processes that do not share a global clock, \emph{partial synchrony} is achieved by clock synchronization that guarantees bounded clock skew among all applications. Existing solutions for distributed runtime verification under partial synchrony against temporal logic specifications are exact but suffer from significant computational overhead. In this paper, we propose a…
▽ More
In distributed systems with processes that do not share a global clock, \emph{partial synchrony} is achieved by clock synchronization that guarantees bounded clock skew among all applications. Existing solutions for distributed runtime verification under partial synchrony against temporal logic specifications are exact but suffer from significant computational overhead. In this paper, we propose an \emph{approximate} distributed monitoring algorithm for Signal Temporal Logic (STL) that mitigates this issue by abstracting away potential interleaving behaviors. This conservative abstraction enables a significant speedup of the distributed monitors, albeit with a tradeoff in accuracy. We address this tradeoff with a methodology that combines our approximate monitor with its exact counterpart, resulting in enhanced efficiency without sacrificing precision. We evaluate our approach with multiple experiments, showcasing its efficacy in both real-world applications and synthetic examples.
△ Less
Submitted 9 August, 2024;
originally announced August 2024.
-
Can AI Understand Our Universe? Test of Fine-Tuning GPT by Astrophysical Data
Authors:
Yu Wang,
Shu-Rui Zhang,
Aidin Momtaz,
Rahim Moradi,
Fatemeh Rastegarnia,
Narek Sahakyan,
Soroush Shakeri,
Liang Li
Abstract:
ChatGPT has been the most talked-about concept in recent months, captivating both professionals and the general public alike, and has sparked discussions about the changes that artificial intelligence (AI) will bring to the world. As physicists and astrophysicists, we are curious about if scientific data can be correctly analyzed by large language models (LLMs) and yield accurate physics. In this…
▽ More
ChatGPT has been the most talked-about concept in recent months, captivating both professionals and the general public alike, and has sparked discussions about the changes that artificial intelligence (AI) will bring to the world. As physicists and astrophysicists, we are curious about if scientific data can be correctly analyzed by large language models (LLMs) and yield accurate physics. In this article, we fine-tune the generative pre-trained transformer (GPT) model by the astronomical data from the observations of galaxies, quasars, stars, gamma-ray bursts (GRBs), and the simulations of black holes (BHs), the fine-tuned model demonstrates its capability to classify astrophysical phenomena, distinguish between two types of GRBs, deduce the redshift of quasars, and estimate BH parameters. We regard this as a successful test, marking the LLM's proven efficacy in scientific research. With the ever-growing volume of multidisciplinary data and the advancement of AI technology, we look forward to the emergence of a more fundamental and comprehensive understanding of our universe. This article also shares some interesting thoughts on data collection and AI design. Using the approach of understanding the universe - looking outward at data and inward for fundamental building blocks - as a guideline, we propose a method of series expansion for AI, suggesting ways to train and control AI that is smarter than humans.
△ Less
Submitted 14 April, 2024;
originally announced April 2024.
-
Decentralized Predicate Detection over Partially Synchronous Continuous-Time Signals
Authors:
Charles Koll,
Anik Momtaz,
Borzoo Bonakdarpour,
Houssam Abbas
Abstract:
We present the first decentralized algorithm for detecting predicates over continuous-time signals under partial synchrony. A distributed cyber-physical system (CPS) consists of a network of agents, each of which measures (or computes) a continuous-time signal. Examples include distributed industrial controllers connected over wireless networks and connected vehicles in traffic. The safety require…
▽ More
We present the first decentralized algorithm for detecting predicates over continuous-time signals under partial synchrony. A distributed cyber-physical system (CPS) consists of a network of agents, each of which measures (or computes) a continuous-time signal. Examples include distributed industrial controllers connected over wireless networks and connected vehicles in traffic. The safety requirements of such CPS, expressed as logical predicates, must be monitored at runtime. This monitoring faces three challenges: first, every agent only knows its own signal, whereas the safety requirement is global and carries over multiple signals. Second, the agents' local clocks drift from each other, so they do not even agree on the time. Thus, it is not clear which signal values are actually synchronous to evaluate the safety predicate. Third, CPS signals are continuous-time so there are potentially uncountably many safety violations to be reported. In this paper, we present the first decentralized algorithm for detecting conjunctive predicates in this setup. Our algorithm returns all possible violations of the predicate, which is important for eliminating bugs from distributed systems regardless of actual clock drift. We prove that this detection algorithm is in the same complexity class as the detector for discrete systems. We implement our detector and validate it experimentally.
△ Less
Submitted 4 August, 2023;
originally announced August 2023.
