Dr. Juraj Hasik | Introduction to PEPS Part II

Value of the Information & Strength of the Argument

The information presented is highly valuable for researchers and advanced students in quantum many-body physics and tensor network methods. The argumentation is solid: the speaker builds the algorithm step-by-step, justifying each step with mathematical reasoning and connecting it to known optimal approximation results from matrix product states. The use of the Ising model as a testbed is effective because its partition function can be exactly translated to a tensor network, making the method’s validity clear. The explanation of how to handle finite-size and infinite systems is thorough, with a clear emphasis on the SVD truncation as a controllable approximation. The lecture also addresses potential questions, clarifying the relationship between the classical Ising model tensor and the PEPS tensor, which strengthens the presentation’s coherence. The main caveat is that the method’s efficiency and accuracy in general PEPS cases are not deeply analyzed, but for an introductory part, the argumentation is convincing.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor of the lecture is high: it provides a self-contained mathematical derivation of the corner transfer matrix algorithm for two-dimensional tensor networks. The source materials are implicit within the tensor network literature, but the speaker does not explicitly cite references during the talk; however, the Isaac Newton Institute seminar page provides links to the event and the institute, which serve as reliable sources. The title ‘Introduction to PEPS Part II’ is appropriate: the content indeed builds on prior knowledge of PEPS and focuses on a specific algorithmic aspect. No extraneous or unverified claims are made; the discussion stays within the established framework of tensor network approximations. The lecture’s coherence and the speaker’s expertise (University of Zurich) contribute to its trustworthiness.

The title accurately reflects the content, which is the second part of an introduction to Projected Entangled Pair States (PEPS), focusing on the corner transfer matrix approach for their approximation.

8/10

The lecture is given by a researcher at the University of Zurich and delivered at the Isaac Newton Institute, a world-leading mathematics institute. The content is mathematically rigorous, with clear derivations and proper referencing to standard tensor network methods. Potential limitations include a narrow focus on a specific algorithm without extensive comparative analysis, but the presentation is technically sound.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 00:00Introduction and outline: the goal is to approximate observables in infinite PEPS using corner transfer matrices.
• 04:00Transcription of the 2D Ising model partition function into a tensor network, defining delta tensors and bond matrices.
• 15:00Construction of corner (C) and edge (T) tensors for finite-size systems, allowing systematic enlargement of the lattice.
• 25:00Discussion of the reduced density matrix and the use of SVD truncation to approximate corner tensors.
• 35:00Summary of the corner transfer matrix algorithm and its connection to general PEPS tensor network contraction.

• Isaac Newton Institute — Main website of the institute hosting the lecture.
• Seminar page for the talk — Official page providing details about the event (MMBW02 European tensor networks school).