- Rating
- 1
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 1
- Soundness
- 3
Read review
Paper summary
The paper considers the problem of choosing a reduced dictionary of codes, out of pre-trained encoders set, such that at inference time every service request (given desired performance, that is, variational rate and distortion ) is answered with an adequate working point. The main method proposes first to allocate each request to one of the existing options (called an “anchor”), and then use classical dynamic programming to choose a dictionary optimize an upper bound on the loss.
Critical strengths
- The main method is not confined to coding problems, and can be used as a method for more general service allocation settings, replacing and with relevant metrics.
- The proposed method is efficient, and has polynomial time complexity.
- Error guarantee and sample complexity analysis are given.
Critical weaknesses
-
The first strength is also the major weakness - the core method is somewhat straightforward, and general. The paper devotes most of its focus to the VAE problem, but there is nothing specific about the VAE problem that makes the routing problem specific.
-
Error analysis, although not completely trivial, is given in terms of the maximal diversity from optimum along atoms, which might not be tight.
-
Overall setting is not well-motivated (see Questions 4 and 5 below).
-
The main text is not clearly written, barely readable and lacks many of the central definitions and the main Algorithm. The Appendix is more carefully edited, but the paper is judged based on the main text.
Questions for authors
-
What are weights w_{n} and how they are chosen. (l.266)
-
In (l.273) “a request can end up closer to an interior atom than to either atom flanking it in D-order”. This central claim remains unclear. please clarify this (i.e. give a concrete example).
-
(l.660) Define h_{n} clearly.
-
In lossy compression, the classical Rate-Distortion tradeoff [R1,R2] relate the maximal tolerable distortion with a minimal rate achieving this distortion at reconstruction. That is, by setting the rate of a scheme, an optimal distortion is automatically set, and vice versa. In your setting, a “request” is defined by the point , which might be non-optimal or non-achievable. Please add motivation for this mode of operation.
-
When traversing a Pareto at inference time, a common practice is an interpolation between pre-trained models. This can be done for example in pixel space [R3], or in the model (parameters) space [R4]. In your method, is it possible to interpolate between endpoints instead of routing the anchor to one of them? How would it affect the deviation from true loss?
[R1] Thomas M. Cover and Joy A. Thomas. Elements of information theory. John Wiley & Sons, 2012.
[R2] Toby Berger. Rate-distortion theory. Wiley Encyclopedia of Telecommunications, 2003.
[R3] Xintao Wang, Ke Yu, Shixiang Wu, Jinjin Gu, Yihao Liu, Chao Dong, Yu Qiao, and Chen Change Loy. ESRGAN: Enhanced super-resolution generative adversarial networks. In Proceedings of the European Conference on Computer Vision (ECCV) Workshops, pages 0–0, 2018.
[R4] Xintao Wang, Ke Yu, Chao Dong, Xiaoou Tang, and Chen Change Loy. Deep network interpolation for continuous imagery effect transition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1692–1701, 2019.