Proof in the time of machines

Andrew Granville

2023-05-03

Abstract:We compare the values associated with (traditional) community based proof verification to those associated with computer proof verification. We propose ways that computer proofs might incorporate successful strategies from human experiences.

Logic in Computer Science,History and Overview,Number Theory

What problem does this paper attempt to address?

Proof in the time of machines

Andrew Granville

DOI: https://doi.org/10.1090/bull/1826

2024-02-15

Bulletin of the American Mathematical Society

Abstract:We are concerned here with the nature of proof and what proof will become in this age of machines. We do so by comparing the values associated with (traditional) community based proof verification to those associated with computer proof verification. We finish by proposing ways that computer proofs might incorporate successful strategies from human experiences.

mathematics
Accepted proofs: Objective truth, or culturally robust

Andrew Granville

2023-05-03

Abstract:How does the mathematical community accept that a given proof is correct? Is objective verification based on explicit axioms feasible, or must the reviewer's experiences and prejudices necessarily come into play? Can automated provers avoid mistakes (as well as experiences and prejudices) to provide objective verification? And can an automated prover's claims be provably verified? We will follow examples of proofs that were found to be flawed, but then corrected (as the proof plan was sufficiently robust), as well as accepted ``proofs'' that turned out to be fundamentally wrong. What does this imply about the desirability of the current community standard for proofs? We will discuss whether mathematical culture is unavoidably part of the acceptance of a proof, no matter how much we try to develop foolproof, objective ``proof systems''.

History and Overview,Algebraic Geometry,Logic,Number Theory
Proof of All: Verifiable Computation in a Nutshell

Mario Alessandro Barbara

DOI: https://doi.org/10.48550/arXiv.1908.02327

2019-08-31

Abstract:Recent advances in the cryptographic field of "Zero-Knowledge Proofs" have sparked a new wave of research, giving birth to many exciting theoretical approaches in the last few years. Such research has often overlapped with the need for private and scalable solutions of Blockchain-based communities, resulting in the first practical implementations of such systems. Many of these innovative constructions have developed in parallel, using different terminologies and evolving into a fragmented ecosystem, calling for their consolidation into the more stable domain of "Verifiable Computation". In this master thesis I propose a unifying Verifiable Computation model for the simplification and efficient comparison of all cryptographic proof systems. I take advantage of this model to analyse innovative technologies (Homomorphic Authenticators, Verifiable Delay Functions) which developed into their own specialised domains, and I attempt to make them more accessible for newcomers to the field. Furthermore, I expand on the future of Verifiable Computation, Universal proof compilers and "Proofs of All", by approaching the state-of-the-art zk-STARK construction from a more accessible and informal design perspective.

Cryptography and Security
Interactive Proofs for Quantum Computation.

Andrew Chi-Chih Yao

DOI: https://doi.org/10.1007/978-3-540-24587-2_1

2003-01-01

Abstract:It is by now well established that quantum machines can solve certain computational problems much faster than the best algorithms known in the standard Turing machine model. The complexity question of which problems can be feasibly computed by quantum machines has also been extensively investigated in recent years, both in the context of one machine models (quantum polynomial classes) and various flavors of multi-machine models (single and multiple prover quantum interactive proofs). In this talk we examine the more general (but less theoretically investigated) question of which quantum states can be feasibly computed. Specifically, we will focus on the question of what quantum states can be generated by quantum interactive proofs. We will show that several classical interactive proof theorems have analogs in such models. For example, we show that any quantum state computable in quantum polynomial space has a 2-prover quantum interactive proof. Open questions will be discussed.
Truth, Proof, and Reproducibility: There's no counter-attack for the codeless

Charles T. Gray,Ben Marwick

DOI: https://doi.org/10.48550/arXiv.1907.05947

2019-07-11

Abstract:Current concerns about reproducibility in many research communities can be traced back to a high value placed on empirical reproducibility of the physical details of scientific experiments and observations. For example, the detailed descriptions by 17th century scientist Robert Boyle of his vacuum pump experiments are often held to be the ideal of reproducibility as a cornerstone of scientific practice. Victoria Stodden has claimed that the computer is an analog for Boyle's pump -- another kind of scientific instrument that needs detailed descriptions of how it generates results. In the place of Boyle's hand-written notes, we now expect code in open source programming languages to be available to enable others to reproduce and extend computational experiments. In this paper we show that there is another genealogy for reproducibility, starting at least from Euclid, in the production of proofs in mathematics. Proofs have a distinctive quality of being necessarily reproducible, and are the cornerstone of mathematical science. However, the task of the modern mathematical scientist has drifted from that of blackboard rhetorician, where the craft of proof reigned, to a scientific workflow that now more closely resembles that of an experimental scientist. So, what is proof in modern mathematics? And, if proof is unattainable in other fields, what is due scientific diligence in a computational experimental environment? How do we measure truth in the context of uncertainty? Adopting a manner of Lakatosian conversant conjecture between two mathematicians, we examine how proof informs our practice of computational statistical inquiry. We propose that a reorientation of mathematical science is necessary so that its reproducibility can be readily assessed.

History and Overview,Other Statistics
Proofs about Network Communication: For Humans and Machines

Wolfgang Jeltsch,Javier Díaz

DOI: https://doi.org/10.4204/EPTCS.383.1

2023-08-21

Abstract:Many concurrent and distributed systems are safety-critical and therefore have to provide a high degree of assurance. Important properties of such systems are frequently proved on the specification level, but implementations typically deviate from specifications for practical reasons. Machine-checked proofs of bisimilarity statements are often useful for guaranteeing that properties of specifications carry over to implementations. In this paper, we present a way of conducting such proofs with a focus on network communication. The proofs resulting from our approach are not just machine-checked but also intelligible for humans.

Logic in Computer Science,Distributed, Parallel, and Cluster Computing
AfterMath: The Work of Proof in the Age of Human–Machine Collaboration

Stephanie Dick

DOI: https://doi.org/10.1086/661623

2011-09-01

Isis

Abstract:During the 1970s and 1980s, a team of Automated Theorem Proving researchers at the Argonne National Laboratory near Chicago developed the Automated Reasoning Assistant, or AURA, to assist human users in the search for mathematical proofs. The resulting hybrid humans+AURA system developed the capacity to make novel contributions to pure mathematics by very untraditional means. This essay traces how these unconventional contributions were made and made possible through negotiations between the humans and the AURA at Argonne and the transformation in mathematical intuition they produced. At play in these negotiations were experimental practices, nonhumans, and nonmathematical modes of knowing. This story invites an earnest engagement between historians of mathematics and scholars in the history of science and science studies interested in experimental practice, material culture, and the roles of nonhumans in knowledge making.

history & philosophy of science
Machine-Checked Proofs For Realizability Checking Algorithms

Andreas Katis,Andrew Gacek,Michael W. Whalen

DOI: https://doi.org/10.48550/arXiv.1502.01292

2015-11-18

Abstract:Virtual integration techniques focus on building architectural models of systems that can be analyzed early in the design cycle to try to lower cost, reduce risk, and improve quality of complex embedded systems. Given appropriate architectural descriptions, assume/guarantee contracts, and compositional reasoning rules, these techniques can be used to prove important safety properties about the architecture prior to system construction. For these proofs to be meaningful, each leaf-level component contract must be realizable; i.e., it is possible to construct a component such that for any input allowed by the contract assumptions, there is some output value that the component can produce that satisfies the contract guarantees. We have recently proposed (in [1]) a contract-based realizability checking algorithm for assume/guarantee contracts over infinite theories supported by SMT solvers such as linear integer/real arithmetic and uninterpreted functions. In that work, we used an SMT solver and an algorithm similar to k-induction to establish the realizability of a contract, and justified our approach via a hand proof. Given the central importance of realizability to our virtual integration approach, we wanted additional confidence that our approach was sound. This paper describes a complete formalization of the approach in the Coq proof and specification language. During formalization, we found several small mistakes and missing assumptions in our reasoning. Although these did not compromise the correctness of the algorithm used in the checking tools, they point to the value of machine-checked formalization. In addition, we believe this is the first machine-checked formalization for a realizability algorithm.

Software Engineering
Computer-Aided Verification of P/NP Proofs: A Survey and Discussion

Stefan Rass,Max-Julian Jakobitsch,Stefan Haan,Moritz Hiebler

DOI: https://doi.org/10.1109/access.2024.3355540

IF: 3.9

2024-02-03

IEEE Access

Abstract:We survey a collection of proofs towards equality, inequality, or independence of the relation of P to NP. Since the problem has attracted much attention from experts, amateurs, and in-betweens, this work is intended as a pointer into directions to enable a "self-assessment" of ideas laid out by people interested in the problem. To this end, we identify the most popular approaches to proving equality, inequality, or independence. Since the latter category appears to be without any attempts to follow the necessary proof strategies, we devote a Section to an intuitive outline of how independence proofs would work. In the other cases of proving equality or inequality, known barriers like (affine) relativization, algebrization, and others are to be avoided. The most important and powerful technique available in this regard is a formalization of arguments in automated proof assistants. This allows an objective self-check of a proof before presenting it to the scientific community.

computer science, information systems,telecommunications,engineering, electrical & electronic
Interactive Machines and Its Implication in Scientific Realism

Shuai Han

DOI: https://doi.org/10.2991/assehr.k.211220.485

2022-01-01

Abstract:Since the tech boom in the early 2000s, computer science, especially the study of algorithms and mathematics, has been prevalent to satisfy new needs for efficiency and convenience. The founding father of modern computer science, Alan Turing, proposes the foundation of modern computability theories known as the Turing Machine. This article is dedicated to exploring a variant of Turing’s mathematical model, where the inputs become infinite and empirical. In a way, it is also an exploration of whether theoretically a true interactive machine can be achieved through a finite number of noninteractive machines. To prove it possible, this article utilizes methods such as researching from established papers, proof by deduction, and mathematical induction. In terms of articles, the main source is JSTOR. As the proof goes, it is mathematically proven in this essay that in terms of generic case K, an Interactive Machine A(K) shares the same explanatory equivalence with a finite number of Turing Machines T(K). It is said that A maps onto T.
Interactive Proofs with Quantum Finite Automata

Harumichi Nishimura,Tomoyuki Yamakami

DOI: https://doi.org/10.1016/j.tcs.2014.11.030

2014-08-16

Abstract:Following an early work of Dwork and Stockmeyer on interactive proof systems whose verifiers are two-way probabilistic finite automata, the authors initiated in 2004 a study on the computational power of quantum interactive proof systems whose verifiers are particularly limited to quantum finite automata. As a follow-up to the authors' early journal publication [J. Comput. System Sci., vol.75, pp.255-269, 2009], we further investigate the quantum nature of interactions between provers and verifiers by studying how various restrictions on quantum interactive proof systems affect the language recognition power of the proof systems. In particular, we examine three intriguing restrictions that (i) provers always behave in a classical fashion, (ii) verifiers always reveal to provers the information on next moves, and (iii) the number of interactions between provers and verifiers is bounded.

Quantum Physics,Computational Complexity,Formal Languages and Automata Theory
Mathematical reasoning and the computer

Kevin Buzzard

DOI: https://doi.org/10.1090/bull/1833

2024-02-15

Bulletin of the American Mathematical Society

Abstract:Computers have already changed the way that humans do mathematics: they enable us to compute efficiently. But will they soon be helping us to reason ? And will they one day start reasoning themselves? We give an overview of recent developments in neural networks, computer theorem provers, and large language models.

mathematics
Strange new universes: Proof assistants and synthetic foundations

Michael Shulman

DOI: https://doi.org/10.1090/bull/1830

2024-02-15

Bulletin of the American Mathematical Society

Abstract:Existing computer programs called proof assistants can verify the correctness of mathematical proofs but their specialized proof languages present a barrier to entry for many mathematicians. Large language models have the potential to lower this barrier, enabling mathematicians to interact with proof assistants in a more familiar vernacular. Among other advantages, this may allow mathematicians to explore radically new kinds of mathematics using an LLM-powered proof assistant to train their intuitions as well as ensure their arguments are correct. Existing proof assistants have already played this role for fields such as homotopy type theory.

mathematics
Unconditional proofs of quantumness between small-space machines

A. C. Cem Say,M. Utkan Gezer

2024-12-04

Abstract:A proof of quantumness is a protocol through which a classical machine can test whether a purportedly quantum device, with comparable time and memory resources, is performing a computation that is impossible for classical computers. Existing approaches to provide proofs of quantumness depend on unproven assumptions about some task being impossible for machines of a particular model under certain resource restrictions. We study a setup where both devices have space bounds $\mathit{o}(\log \log n)$. Under such memory budgets, it has been unconditionally proven that probabilistic Turing machines are unable to solve certain computational problems. We formulate a new class of problems, and show that these problems are polynomial-time solvable for quantum machines, impossible for classical machines, and have the property that their solutions can be "proved" by a small-space quantum machine to a classical machine with the same space bound. These problems form the basis of our newly defined protocol, where the polynomial-time verifier's verdict about the tested machine's quantumness is not conditional on an unproven weakness assumption.

Computational Complexity,Formal Languages and Automata Theory,Quantum Physics
Interactive Proofs For Quantum Computations

Dorit Aharonov,Michael Ben-Or,Elad Eban

DOI: https://doi.org/10.48550/arXiv.0810.5375

2008-11-18

Abstract:The widely held belief that BQP strictly contains BPP raises fundamental questions: Upcoming generations of quantum computers might already be too large to be simulated classically. Is it possible to experimentally test that these systems perform as they should, if we cannot efficiently compute predictions for their behavior? Vazirani has asked: If predicting Quantum Mechanical systems requires exponential resources, is QM a falsifiable theory? In cryptographic settings, an untrusted future company wants to sell a quantum computer or perform a delegated quantum computation. Can the customer be convinced of correctness without the ability to compare results to predictions? To answer these questions, we define Quantum Prover Interactive Proofs (QPIP). Whereas in standard Interactive Proofs the prover is computationally unbounded, here our prover is in BQP, representing a quantum computer. The verifier models our current computational capabilities: it is a BPP machine, with access to few qubits. Our main theorem can be roughly stated as: "Any language in BQP has a QPIP, and moreover, a fault tolerant one". We provide two proofs. The simpler one uses a new (possibly of independent interest) quantum authentication scheme (QAS) based on random Clifford elements. This QPIP however, is not fault tolerant. Our second protocol uses polynomial codes QAS due to BCGHS, combined with quantum fault tolerance and multiparty quantum computation techniques. A slight modification of our constructions makes the protocol "blind": the quantum computation and input are unknown to the prover. After we have derived the results, we have learned that Broadbent at al. have independently derived "universal blind quantum computation" using completely different methods. Their construction implicitly implies similar implications.

Quantum Physics
A review and prospect of readable machine proofs for geometry theorems

Jianguo Jiang,Jingzhong Zhang

DOI: https://doi.org/10.1007/s11424-012-2048-3

2012-01-01

Abstract:After half a century research, the mechanical theorem proving in geometries has become an active research topic in the automated reasoning field. This review involves three approaches on automated generating readable machine proofs for geometry theorems which include search methods, coordinate-free methods, and formal logic methods. Some critical issues about these approaches are also discussed. Furthermore, the authors propose three further research directions for the readable machine proofs for geometry theorems, including geometry inequalities, intelligent geometry softwares and machine learning.
Euclid After Computer Proof-checking

Michael Beeson

DOI: https://doi.org/10.1080/00029890.2022.2069985

2021-03-17

Abstract:Euclid pioneered the concept of a mathematical theory developed from axioms by a series of justified proof steps. From the outset there were critics and improvers. In this century the use of computers to check proofs for correctness sets a new standard of rigor. How does Euclid stand up under such an examination? And what does the exercise have to teach us about geometry, mathematical foundations, and the relation of logic to truth?

History and Overview
Machines, logic and quantum physics

David Deutsch,Artur Ekert,Rossella Lupacchini

DOI: https://doi.org/10.2307/421056

2000-01-01

Bulletin of Symbolic Logic

Abstract:Though the truths of logic and pure mathematics are objective and independent of any contingent facts or laws of nature, our knowledge of these truths depends entirely on our knowledge of the laws of physics. Recent progress in the quantum theory of computation has provided practical instances of this, and forces us to abandon the classical view that computation, and hence mathematical proof, are purely logical notions independent of that of computation as a physical process. Henceforward, a proof must be regarded not as an abstract object or process but as a physical process, a species of computation, whose scope and reliability depend on our knowledge of the physics of the computer concerned.
QED at Large: A Survey of Engineering of Formally Verified Software

Talia Ringer,Karl Palmskog,Ilya Sergey,Milos Gligoric,Zachary Tatlock

DOI: https://doi.org/10.1561/2500000045

2020-03-14

Abstract:Development of formal proofs of correctness of programs can increase actual and perceived reliability and facilitate better understanding of program specifications and their underlying assumptions. Tools supporting such development have been available for over 40 years, but have only recently seen wide practical use. Projects based on construction of machine-checked formal proofs are now reaching an unprecedented scale, comparable to large software projects, which leads to new challenges in proof development and maintenance. Despite its increasing importance, the field of proof engineering is seldom considered in its own right; related theories, techniques, and tools span many fields and venues. This survey of the literature presents a holistic understanding of proof engineering for program correctness, covering impact in practice, foundations, proof automation, proof organization, and practical proof development.

Logic in Computer Science,Programming Languages
Perspectives for proof unwinding by programming languages techniques

Danko Ilik

DOI: https://doi.org/10.48550/arXiv.1605.09177

2016-05-30

Abstract:In this chapter, we propose some future directions of work, potentially beneficial to Mathematics and its foundations, based on the recent import of methodology from the theory of programming languages into proof theory. This scientific essay, written for the audience of proof theorists as well as the working mathematician, is not a survey of the field, but rather a personal view of the author who hopes that it may inspire future and fellow researchers.

Logic,Logic in Computer Science,Programming Languages

Proof in the time of machines

Proof in the time of machines

Accepted proofs: Objective truth, or culturally robust

Proof of All: Verifiable Computation in a Nutshell

Interactive Proofs for Quantum Computation.

Truth, Proof, and Reproducibility: There's no counter-attack for the codeless

Proofs about Network Communication: For Humans and Machines

AfterMath: The Work of Proof in the Age of Human–Machine Collaboration

Machine-Checked Proofs For Realizability Checking Algorithms

Computer-Aided Verification of P/NP Proofs: A Survey and Discussion

Interactive Machines and Its Implication in Scientific Realism

Interactive Proofs with Quantum Finite Automata

Mathematical reasoning and the computer

Strange new universes: Proof assistants and synthetic foundations

Unconditional proofs of quantumness between small-space machines

Interactive Proofs For Quantum Computations

A review and prospect of readable machine proofs for geometry theorems

Euclid After Computer Proof-checking

Machines, logic and quantum physics

QED at Large: A Survey of Engineering of Formally Verified Software

Perspectives for proof unwinding by programming languages techniques