Consequences of Randomized Reductions from SAT to Time-Bounded Kolmogorov Complexity
Bibliographic record
Abstract
A central open question within meta-complexity is that of NP-hardness of problems such as MCSP and MK^{t}P. Despite a large body of work giving consequences of and barriers for NP-hardness of these problems under (restricted) deterministic reductions, very little is known in the setting of randomized reductions. In this work, we give consequences of randomized NP-hardness reductions for both approximating and exactly computing time-bounded and time-unbounded Kolmogorov complexity. In the setting of approximate K^{poly} complexity, our results are as follows. 1) Under a derandomization assumption, for any constant δ > 0, if approximating K^t complexity within n^{δ} additive error is hard for SAT under an honest randomized non-adaptive Turing reduction running in time polynomially less than t, then NP = coNP. 2) Under the same assumptions, the worst-case hardness of NP is equivalent to the existence of one-way functions. Item 1 above may be compared with a recent work of Saks and Santhanam [Michael E. Saks and Rahul Santhanam, 2022], which makes the same assumptions except with ω(log n) additive error, obtaining the conclusion NE = coNE. In the setting of exact K^{poly} complexity, where the barriers of Item 1 and [Michael E. Saks and Rahul Santhanam, 2022] do not apply, we show: 3) If computing K^t complexity is hard for SAT under reductions as in Item 1, then the average-case hardness of NP is equivalent to the existence of one-way functions. That is, "Pessiland" is excluded. Finally, we give consequences of NP-hardness of exact time-unbounded Kolmogorov complexity under randomized reductions. 4) If computing Kolmogorov complexity is hard for SAT under a randomized many-one reduction running in time t_R and with failure probability at most 1/(t_R)^16, then coNP is contained in non-interactive statistical zero-knowledge; thus NP ⊆ coAM. Also, the worst-case hardness of NP is equivalent to the existence of one-way functions. We further exploit the connection to NISZK along with a previous work of Allender et al. [Eric Allender et al., 2023] to show that hardness of K complexity under randomized many-one reductions is highly robust with respect to failure probability, approximation error, output length, and threshold parameter.
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 | 0.009 | 0.060 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.002 | 0.008 |
| Scholarly communication | 0.004 | 0.012 |
| Open science | 0.004 | 0.008 |
| Research integrity | 0.003 | 0.014 |
| Insufficient payload (model declined to judge) | 0.008 | 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".