On the Design of Symmetric Entropy-constrained Multiple Description Scalar Quantizer with Linear Joint Decoders
Bibliographic record
Abstract
This paper addresses the design of symmetric entropy-constrained multiple description scalar quantizers (EC-MDSQ) with linear joint decoders, i.e., where some of the decoders compute the reconstruction by averaging the reconstructions of individual descriptions. Thus, the use of linear decoders reduces the space complexity at the decoder since only a subset of the codebooks needs to be stored. The proposed design algorithm locally minimizes the Lagrangian, which is a weighted sum of the expected distortion and of the side quantizers' rates. The algorithm is inspired by the EC-MDSQ design algorithm proposed by Vaishampayan and Domaszewicz, and it is adapted from two to K descriptions. Differently from the aforementioned work, the optimization of the reconstruction values can no longer be performed separately at the decoder optimization step. Interestingly, we show that the problem is a convex quadratic optimization problem, which can be efficiently solved. Moreover, the generalization of the encoder optimization step from two to K descriptions increases drastically the amount of computations. We show how to exploit the special form of the cost function conferred by the linear joint decoders to significantly reduce the time complexity at this step. We compare the performance of the proposed design with multiple description lattice vector quantizers (MDLVQ) and with the multiple description scheme based on successive refinement and unequal erasure protection (UEP). Our experiments show that the proposed approach outperforms MDLVQ with dimension 1 quantization, as expected. Additionally, when more codebooks are added our scheme even beats MDLVQ with quantization dimension approaching ∞, for rates sufficiently high. Furthermore, the proposed approach is also superior to UEP with dimension 1 quantization when the rates are low.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".