Due to COVID-19 outbreak across the world, we need to move our theory seminars online to ZOOM. We are staying on track with the schedule and Yoriyuki Yamagata will give his talk tomorrow at 2pm.
Topic: Falsification of Cyber-Physical Systems Using Deep Reinforcement Learning.
Abstract: “Falsification” is a method to find a system input or parameter (counter-example) which causes a behavior violating a given specification (usually given by metric or signal temporal logic). Because the correctness of a complex CPS is difficult to be proven, falsification is more practical approach than full verification. A counter-example found by falsification can be used for debugging and testing. Failure of falsification does not generally mean the correctness of the system, but suggests it in some degree. “Robustness guided falsification” is an approach of falsification. “Robustness” is a numerical measure of how robustly a formula holds. If robustness becomes negative, the formula is false. Therefore, minimizing robustness can lead falsification of a formula.
In this talk, we introduce a method to recast robustness guided falsification to a “reinforcement learning problem”. Reinforcement learning is a machine learning technique in which an agent finds a law of an interacting environment and maximizes a reward. We implement our method using “deep reinforcement leaning”, in which deep neural networks are used, and present a case study to explore its effectiveness. (This work is a collaboration with Shuang Liu, Takumi Akazaki, Yihai Duan, Jianye Hao)
Senior Researcher, Software Analytics Research Group, Information Technology Reseaerch Institute, AIST
We remain hopeful that our conference will go ahead as planned. However, due to the uncertainty created by COVID-19, we are putting into place contingency plans.
Swansea University has procured a licence for a robust video-conferencing system (ZOOM) which we can use for the conference. If it becomes necessary, then a user would need to install this onto their computer. This system is free to install and use, the only cost is for Swansea as the licence holder. Of course, the computer would need to be connected to a webcam and microphone in order to participate fully in the conference – in particular, to deliver a lecture or to ask questions of the speakers.
Anyone who cannot come to Swansea due to the COVID-19 situation would have their fee reimbursed through eventbrite (as you won’t be requiring the catering). However, we really don’t want this to be an incentive to stay home; AlgoUK/BCTCS is first and foremost a net working event, so if it is possible, we are keen to welcome you in person.
We will keep you aware of any developments, and thank you for your patience and understanding as we do our best to ensure AlgoUK/BCTCS is as successful as it can be.
We are glad to welcome Hideki Tsuiki in Swansea again. This Thursday we really enjoyed his talk on “Imaginary Cubes — Mathematics, Puzzle, Art and Education”.
Abstract: Imaginary cubes are three-dimensional objects with square projections in three orthogonal ways just as a cube has. How many different kinds of imaginary cubes can you imagine? In this talk we show that there are 16 kinds of minimal convex imaginary cubes which includes regular tetrahedron, cuboctahedron, and two objects that we call H and T. As we will explain, H and T have a lot of beautiful mathematical properties related to tiling, fractal, and higher-dimensional geometry, and based on these properties, the speaker has designed a puzzle, constructed three-dimensional math-art objects, and used them for educations at various levels from elemental school to graduate schools. In this talk, I will explain mathematics of imaginary cubes and show the activities I have been engaged in. I will carry a couple of copies of the puzzle and some of the math-art objects so that the audience can enjoy them while I am staying in Swansea.
Following the Summer School, we are really proud to host the 2nd Proof Society Workshop. The workshop was an opportunity to listen to a lot of interesting invited and contributed talks on proof theory and various areas of its application:
Adam Wyner: Computational Law – The Case of Autonomous Vehicles Yong Cheng: Exploring the incompleteness phenomenon Matthias Baaz: Towards a Proof Theory for Henkin Quantifiers Sonia Marin: On cut-elimination for non-wellfounded proofs: the case of PDL Gilles Dowek: Logical frameworks, reverse mathematics, and formal proofs translation Benjamin Ralph: What is a combinatorial proof system? William Stirton: Ordinal assignments correlated with notions of reduction Oliver Kullmann: Practical proof theory: practical versions of Extended Resolution Anton Setzer and Ulrich Berger on behalf of Ralph Matthes: Martin Hofmann’s case for non-strictly positive data types – reloaded Laura Crosilla: Philosophy of mathematics and proof theory Takako Nemoto: Recursion Theory in Constructive Mathematics Arno Pauly: Combinatorial principles equivalent to weak induction Antonina Kolokolova: The proof complexity of reasoning over richer domains Joost Joosten: The reduction property revisited Helmut Schwichtenberg: Computational content of proofs
Thanks to all the speaker and participants and we hope to see you all again soon.
Monika Seisenberger (Swansea University) and Lauri Hella (Tampere University, Finland) is chairing the Twenteeth International Workshop on Logic and Computational Complexity (LLC’19), which will be held in Patras, Greece, on July 8, 2019, as part of ICALP.
More information is here:
Logic and Computational Complexity
The Twenteeth International Workshop on Logic and Computational Complexity will be held in Patras, Greece, on July 8, 2019, as part of ICALP.
As a part of our Theory Seminars, today we welcomed Ryota Akiyoshi from Waseda University, who gave a talk on Takeuti’s finitism.
Abstract: In this talk, we address several mathematical and philosophical issues of Gaisi Takeuti’s proof theory, who is one of the most distinguished logicians in proof theory after Hilbert and Gentzen. He furthered the realization of Hilbert’s program by formulating Gentzen’s sequent calculus for higher-oder logics, conjecturing that the cut-elimination holds for it (Takeuti’s conjecture), and obtaining several stunning results in the 1950-60’s towards the solution of his conjecture.
This talk consists of two parts. (1) To summarize Takeuti’s background and the argument of the well-ordering proof of ordinals up to ε0 , (2) To evaluate it on philosophical grounds. Also, we will explain several mathematical and philosophical issues to be solved. This is joint work with Andrew Arana.
My research fields are philosophy of mathematics and logic, mathematical logic (proof theory), and theoretical computer science (type theory). The main philosophical question in my research is "what is a proof ?".