Random spanning of a binary vector space Created 2026-10-25 Updated 2026-10-31
Take independent uniform samples from a -dimensional vector space over the two-element field. At current matrix rank , the existing span contains of the possible vectors, so a new sample increases the matrix rank with probability . The mean waiting time for that increase is by the geometric distribution. Adding the waiting times gives . For , the bound follows by separating and using for ; for , no samples are needed.
After samples, failure to span implies that some nonzero linear functional vanishes on all samples. Each fixed nonzero functional does so with probability ; the union bound over the possibilities gives failure less than . This provides both an expected sample count and a fixed-budget high-success version of Simon's algorithm.
Random transposition shuffle Created 2026-10-06 Updated 2026-11-05
Choose two positions independently and uniformly from and swap them. This Markov chain on the symmetric group holds with Probability and assigns Probability to each unordered transposition. Its stationary distribution is uniform, and its ordinary spectral gap is for . The latter follows from the Aldous spectral gap theorem: the single-label continuous-time Markov chain has rate between every pair of positions, whose mean-zero eigenvalues are all .
Random variable Created 2026-09-24 Updated 2026-10-14
A real-valued random variable is a measurable function from a probability space to the real numbers with their Borel sigma-algebra. More general targets give random elements. For every Borel set , measurability ensures that is an event, so the question "does the observed value lie in ?" has a defined probability.
The function is fixed; its input outcome is uncertain. This separates what we measure from how the uncertainty is modeled. A probability distribution is the resulting pushforward measure, . A cumulative distribution function, probability mass function, or probability density function describes that measure in an appropriate setting. The random variable and its probability distribution are different objects: the former also retains its relationship to other measurements on the same sample space.
For two independent uniform dice, let record their results and define and . The 36 ordered pairs are equally likely, but the resulting probability mass functions differ:
For the sum, count the diagonal of the outcome grid having total . For the maximum, subtract the pairs with both results below from the pairs with both results at most . These examples show how a deterministic function transforms a simple input probability distribution into a nonuniform output one.
Figure 1.
Different random variables on the same equally likely dice outcome space
. Details.
The corresponding expected values are and . The first follows immediately from linearity of expectation; the second follows from . Neither number must be an attainable result. An expected value describes an average under the probability measure, not a prediction that the next observation will equal it.
The distinction between a random variable and its probability distribution matters for dependence. and have the same discrete uniform distribution and are independent random variables. and a second copy of also have that same marginal probability distribution, but they always agree. Their joint probability distribution is concentrated on the diagonal; the marginals alone cannot say which relationship holds. This is why a model for several measurements must specify their joint law or an appropriate independence assumption.
Indicator random variables turn events into numbers: equals one on and zero elsewhere, and . This connects event counting with expected value and variance, and lets tools for random variables answer questions about probabilities. More complicated functions of observations can describe totals, waiting times, or estimation errors using the same framework.
Random walk Created 2026-09-24 Updated 2026-10-14
A random walk is a stochastic process formed by accumulating random steps. In its additive form, the increments are independent and identically distributed random variables and the position after steps is
The steps can have more choices than and : they may be integers, real numbers, or vectors. A random walk on a graph instead describes steps between vertices along edges; its available moves can depend on the current vertex.
For example, starting at zero, the steps produce the positions . This is one sample path; other step sequences can have the same endpoint. A simple random walk on the integer line makes this distinction particularly tangible: repeated Bernoulli trials determine both a whole sample path and an endpoint with a binomial distribution. All positions below are dimensionless quantities measured in steps of length one, and counts discrete updates.
The model is useful because it separates an average tendency from accumulated fluctuations. A fair simple symmetric random walk has zero expected value at every time, yet its standard deviation grows: cancellation across many possible sample paths does not keep each walker near zero. The two figures below compare sample paths with exact endpoint probabilities. For the basic additive construction and its independent increments, see MIT, Stochastic Processes I, section 2.
Rank-one dephasing Created 2026-10-06 Updated 2026-11-05
In an orthonormal basis, rank-one dephasing is the quantum channel . It removes off-diagonal entries while preserving the basis probabilities. Its output Von Neumann entropy is their Shannon entropy, and the relative-entropy identity for rank-one dephasing quantifies the entropy increase.
Rao-Blackwellization of a multinomial probability product Created 2026-10-07 Updated 2026-11-05
For independent labels with three type probabilities, conditioning the indicator of an ordered initial triple on the multinomial distribution type counts gives the displayed unbiased estimator. The Rao-Blackwell theorem reduces its variance strictly when all three probabilities are positive. At a boundary parameter with a missing type, both variances vanish.
Rare-event collaboration size at half power Created 2026-10-14 Updated 2026-11-05
Suppose each collaborating centre contributes independently randomized participants per arm with specified event probabilities. At 50% statistical power, the usual normal-approximation alternative quantile is zero. The required size per arm is , with for a two-sided test or for a one-sided test. Divide by the per-centre arm size and round up. Endpoints must use the same time origin and follow-up window, and cluster dependence requires an additional adjustment.
Rare-event discovery versus inverse-moment precision separates seeing an influential thin region from estimating its contribution to inferred density. In a two-point thickness model, let a rare thickness occur with probability , with reciprocal thickness , while common regions have reciprocal thickness . For independent observations, the rare count has a binomial distribution. The chance of at least one rare observation is
To make this at least , choose . For stipulated and , the smallest integer is 29,956. This planning statement assumes the specified rare-event probability; it does not estimate that probability from a sample that missed the event.[ref]
The sampling distribution of the estimated inverse moment answers another question. With the observed rare count,
Thus a target relative standard uncertainty for the inverse moment requires
This is a contribution-weighted precision target for inverse moments: the required count depends on both rarity and the leverage of the rare value.
Use the original model , and . Here and . Consequently
At the 95%-discovery count, this relative standard uncertainty is still about 28.9%. Reaching 10% for the inverse moment needs 249,975 observations in this model, while reaching 10% for the rare fraction itself needs 999,900. The factor of four arises because the rare excess accounts for half the total inverse moment, so its relative fluctuation is diluted in the total.
These are exact variance calculations under a stipulated distribution, not confidence-coverage guarantees or measured material properties. Spatially clustered thin regions reduce the effective independent sample count. A credible physical lower-thickness bound or better resolved imaging can change the model more substantially than merely extending a simulation with an unsupported tail.
Rational Decisions by I. J. Good Created 2026-10-13 Updated 2026-11-05
Published in the Journal of the Royal Statistical Society, Series B, volume 14, number 1, pp. 107–114.
Cited passage: the published summary describes encouraging accurate probability estimates and a connection with information theory.
Rayleigh distribution Created 2026-10-06 Updated 2026-10-18
The Rayleigh distribution is the probability distribution of the Euclidean norm of a two-dimensional random vector with independent zero-mean normal random variables of equal variance as components. If are independent, then
Here is the standard deviation of each component, not of the radius . At this gives the unit-scale probability density function .[ref]
This geometry turns two signed errors into one nonnegative distance. For example, suppose horizontal and vertical positioning errors are independent random variables. Their means both vanish, but the distance from the intended point is usually positive. The Rayleigh distribution with gives the probability of landing inside any centered disk. These are chosen model assumptions, rather than measurements of a particular instrument.
Figure 1.
The center maximizes planar density, while a positive radius maximizes radial density
. Original calculated model with . Left: joint probability density per unit area, with an annulus from to . Right: probability density function per unit radius; the shaded area is the probability of that same annulus. A single point and a single circle both have probability zero. Details.
An apparently paradoxical feature is that the most likely location per unit area is the origin, while the radial mode is . A narrow annulus farther out has more area. Near zero that growing circumference outweighs the falling planar density; farther out the exponential function wins. Distinguishing densities measured against different elements of area or length resolves the paradox.
For , integration gives the survival function and cumulative distribution function
Solving gives the quantile function
[ref]
For the chosen positioning model ,
The disk of radius one component standard deviation therefore contains about 39.35% of outcomes; the familiar one-dimensional normal interval percentage does not apply to a two-dimensional disk. The 95% and 99% radial thresholds are
A circle defined by itself still has probability zero; its enclosed disk carries the stated probability.
Figure 1.
Radial coverage uses disks rather than one-dimensional normal intervals
. Original chosen model: each independent centered coordinate has standard deviation . Concentric disks have radii , and the calculated 95% threshold . The curve gives exact model coverage, not empirical counts. Details.
The closed-form quantile function also gives inverse transform sampling: if has uniform distribution on , then has Rayleigh distribution. Using instead gives the same law. Endpoints are excluded so that the logarithm stays finite.
Razborov–Rudich natural-proofs barrier Created 2026-09-28 Updated 2026-11-05
The Razborov–Rudich natural-proofs barrier connects lower bounds in circuit complexity with the security of pseudorandom function families. Suppose a property of Boolean functions is constructive, large and useful against a class containing the relevant pseudorandom function family.
Query all inputs of a candidate Boolean function, then run the property's membership test on its truth table. Constructivity makes this take time. Largeness gives acceptance probability at least for a uniformly random Boolean function. Usefulness gives rejection for all functions in the chosen small-circuit family at sufficiently large lengths. The property therefore supplies a statistical distinguisher.
A security assumption must cover that running time and distinguishing advantage, at the chosen relation between input length and key length. Security only against polynomial-time algorithms in does not cover a full truth table test. Razborov and Rudich use a stronger quantitative hardness assumption to obtain the barrier against P/poly. The conclusion restricts a form of lower-bound proof under the assumption; it does not settle whether P equals NP.[ref]
Reaction-limited polymerization Brownian ratchet Created 2026-10-14 Updated 2026-11-05
When load-gap equilibration is fast compared with chemical events, an opposing force gives an exponential gap distribution. The probability of room for an addition is . With unloaded addition rate and removal rate , assumed force-independent except for this steric gating, the velocity is the displayed expression. For , stall occurs at . Concentration is included in if the on-rate constant is bimolecular.
A reactive strategy cooperates with probability after the opponent cooperated and probability after the opponent defected. It ignores its own previous action. Independent behavioural randomization gives a four-state joint-action Markov chain. The reduced marginal recurrences depend on the prescribed initial action distribution.
Recessive affected-sib-pair likelihood Created 2026-10-14 Updated 2026-11-05
For unaffected genetic carriers of a disease with complete penetrance and recessive inheritance, suppose parental marker-disease haplotype phases are equally likely and the marker is fully informative. Write and , where is the recombination fraction. Conditional on sharing a parental marker copy, the probability that both siblings inherit that parent's disease allele is ; conditional on inheriting different marker copies it is . Independent paternal and maternal transmissions therefore give . Conditional instead on both siblings being affected, the IBD probabilities are , where .
// Target: biology.bigb
Recessive inheritance Created 2026-10-13 Updated 2026-11-05
In a fully penetrant autosomal recessive model with normal allele and disease allele , is affected and are unaffected. A carrier cross has offspring probabilities for , respectively, from the two independent parental meioses.
For normalized linearly independent pure states, their Gram matrix is positive definite. The displayed reciprocal vectors satisfy . The operator has nonzero eigenvalues equal to those of : use the polar factorization of the matrix whose columns are . Choosing gives the POVM effects and . Each conclusive outcome has probability on its own state and zero on every other input. The inconclusive effect preserves positivity, so the construction is a physical measurement in quantum mechanics.
Recombination fraction 2026-10-13
The recombination fraction is the probability that a transmitted gamete carries a recombinant haplotype between two specified genetic loci, relative to the parent's known haplotype phase. Write it as here; the symbol is also common in genetic linkage analysis. In the ordinary symmetric two-locus model, , with for complete genetic linkage and for independent assortment.
For phase-known genotype , fair Mendelian segregation and equal reciprocal gamete probabilities give
For example, 1,000 independent informative transmissions with observed counts in this order have empirical recombination fraction . Those illustrative counts match the expected values in the toy model, but actual counts fluctuate. The genetic linkage diagram shows both the parental and recombinant classes, so their separate probabilities are not confused with the total .
The recombination fraction measures the probability of an odd number of origin switches along the transmitted chromatid. It is neither the expected value of the exchange count nor the probability of at least one exchange: even positive counts can be invisible at the two endpoints. Converting an observed recombination fraction to a genetic map distance therefore needs an explicit model such as the Haldane mapping function. The distinction between transmitted-chromatid counts and four-chromatid exchange counts is discussed in Weeks, Tang and Kwon (2009), Casares' map function: No need for a 'corrected' Haldane's map function, Genetica 135, 305–307.
Record value Created 2026-10-20 Updated 2026-11-05
An upper record value in a sequence is an observation strictly larger than all its predecessors; the first observation is a record by convention. For , the strict upper record values are at times . A tied maximum is not a new strict record. For independent and identically distributed continuous observations, the th observation is a record with probability , since each position is equally likely to hold the maximum of the first values. By linearity of expectation, the expected number of records by time is the harmonic number .
Recovery of a continuous Brownian integrand from short increments Created 2026-10-05 Updated 2026-10-06
For a bounded continuous adapted process and a Brownian motion , converges in probability to as . The Itô isometry makes the error numerator small relative to ; the normal distribution of the denominator controls its chance of being too close to zero.