SciPost Submission Page
Rapid Measurements and Phase Transition Detections Made Simple by AC-GANs
by Jiewei Ding, Ho-Kin Tang, Wing Chi Yu
This is not the latest submitted version.
This Submission thread is now published as
Submission summary
| Authors (as registered SciPost users): | Wing Chi Yu |
| Submission information | |
|---|---|
| Preprint Link: | scipost_202401_00034v1 (pdf) |
| Date submitted: | 2024-01-25 18:10 |
| Submitted by: | Yu, Wing Chi |
| Submitted to: | SciPost Physics |
| Ontological classification | |
|---|---|
| Academic field: | Physics |
| Specialties: |
|
| Approach: | Computational |
Abstract
In recent years, the use of end-to-end neural networks to analyze Monte Carlo data has received a lot of attention. However, the application of non-end-to-end generative ad- versarial neural networks is less explored. Here, we study classical many-body systems using generative adversarial neural networks. We use the conditional generative adversarial network with an auxiliary classifier (AC-GAN) and introduce self-attention layers into the generator, enabling the model to learn the distribution of two-dimensional XY model spin configurations as well as the physical quantities. By applying the symmetry of the systems, we further find that AC-GAN can be trained with a very small raw dataset, allowing us to obtain reliable measurements in the model that requires a large sample size, e.g. the large-sized 2D XY model and the 3D Heisenberg model. We also find that it is possible to quantify the distribution changes that occur in the configuration of the models during phase transitions and locate the phase transition points by AC-GAN.
Current status:
Reports on this Submission
Report #2 by Anonymous (Referee 3) on 2024-3-25 (Invited Report)
- Cite as: Anonymous, Report on arXiv:scipost_202401_00034v1, delivered 2024-03-25, doi: 10.21468/SciPost.Report.8771
Strengths
1-well written and good/clear presentation
2-addresses important question
3-valuable results
Weaknesses
1-a few minor point might require clarification, see the detailed report below
Report
In the manuscript with title “Rapid Measurements and Phase Transition Detection Made Simple by AC-GANs”, Ding, Tang, and Yu study how conditional generative adversarial networks (GANs) with auxiliary classifiers can be used to sample from classical spin models. The models studied are the two-dimensional XY model as well as the Heisenberg model in three spatial dimensions. They find that their method outperforms related previous GAN-based approaches. They further discuss how performing symmetry transformations might be used to obtain better results for small number of samples and also present a way of detecting phase transitions.
I think the manuscript is very well written and presented. It also addresses an important question – how to improve sampling algorithms using generative models – and, as far as I can see, seems to add a valuable contribution. As such, I recommend publication of the work in SciPost physics, in principle. However, I first would like to ask the authors to address the following points:
1) The authors only use simple observables to evaluate the performance of the models. If possible, it could be worthwhile to include other measures such as earth mover distances of the distributions.
2) Did the authors test whether their approach suffers from mode collapse – a well-known issue of GANs?
3) I do not fully understand in which sense using Eq. (7) is beneficial compared to other ways people have studied for the detection of phase transitions. It seems, one needs both MCMC data and train a GAN. Relatedly, in the last paragraph before the conclusion starts, the authors write “It is also different from other machine learning methods in detecting phase transitions where the machine works as a black-box. Our method has a physical interpretation that the phase transition is accompanied by the change in the configuration distribution.” I likely do not understand their approach well enough, but I am not so sure in which sense their approach is less of a black-box than other machine-learning techniques studied before and I think previous approaches are also based on changes in the configuration distributions, see, for instance, Phys. Rev. E 99, 062107 (2019) or also that in Ref. 32.
4) I also noticed a few typos: (i) page 2, “To let the Generator performs …”, (ii) page 3, “… besides the random matrix z, constrains are added …”, and (iii) in the caption of Fig. 2, it seems that column and rows are swapped.
Report #1 by Anonymous (Referee 4) on 2024-3-3 (Invited Report)
- Cite as: Anonymous, Report on arXiv:scipost_202401_00034v1, delivered 2024-03-03, doi: 10.21468/SciPost.Report.8656
Report
Unfortunatly, non of the stated expectations of SciPost physics is matched:
Expectations (at least one required) - the paper must:
Detail a groundbreaking theoretical/experimental/computational discovery;
Present a breakthrough on a previously-identified and long-standing research stumbling block;
Open a new pathway in an existing or a new research direction, with clear potential for multipronged follow-up work;
Provide a novel and synergetic link between different research areas.
-------------------------
Thus, although results seem consistent and the research done has value, I believe this article is not suitable to publish in SciPost physics.
Many of the concepts and or techniques used in the present article are equal or small modifications of https://www.scipost.org/SciPostPhys.11.2.043/pdf
Moreover, the clarity of the written English is far from acceptable, and more details of the network architectures should be added.