Formal probabilistic analysis using theorem proving
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Probabilistic analysis is a tool of fundamental importance to virtually all scientists and engineers as they often have to deal with systems that exhibit random or unpredictable elements. Traditionally, computer simulation techniques are used to perform probabilistic analysis. However, they provide less accurate results and cannot handle large-scale problems due to their enormous computer processing time requirements. To overcome these limitations, this thesis proposes to perform probabilistic analysis by formally specifying the behavior of random systems in higher-order logic and use these models for verifying the intended probabilistic and statistical properties in a computer based theorem prover. The analysis carried out in this way is free from any approximation or precision issues due to the mathematical nature of the models and the inherent soundness of the theorem proving approach. The thesis mainly targets the two most essential components for this task, i.e., the higher-order-logic formalization of random variables and the ability to formally verify the probabilistic and statistical properties of these random variables within a theorem prover. We present a framework that can be used to formalize and verify any continuous random variable for which the inverse of the cumulative distribution function can be expressed in a closed mathematical form. Similarly, we provide a formalization infrastructure that allows us to formally reason about statistical properties, such as mean, variance and tail distribution bounds, for discrete random variables. In order to in illustrate the practical effectiveness of the proposed approach, we consider the probabilistic analysis of three examples: the Coupon Collector's problem, the roundoff error in a digital processor and the Stop-and-Wait protocol. All the above mentioned work is conducted using the HOL theorem prover.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.004 | 0.007 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.000 | 0.002 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it