Absolutely summable sequence space Created 2026-10-06 Updated 2026-10-14
The Banach space consists of absolutely summable scalar sequences, with norm . Its continuous dual is , under the pairing . Finitely supported sequences are norm dense, which identifies a linear functional from its values on coordinate vectors. The space of sequences converging to zero inside this dual is a norming subspace of infinite codimension.
For example, belongs to the absolutely summable sequence space and has norm one, by the geometric series formula. Merely tending to zero is insufficient: tends to zero but its series of absolute values diverges. Thus this sequence space records total absolute size, rather than just the eventual size of individual coordinates. Its completeness ensures that a Cauchy sequence in this total-error norm has an absolutely summable limit.[ref]
https://people.clas.ufl.edu/mjury/files/6617notes30nov2015.pdf. Michael Jury, MAA6617 course notes, section 20.1(b): Banach sequence spaces
Absorption time 2026-10-25
For a discrete-time Markov chain and a set of absorbing states, its absorption time is the hitting time . It is zero when the initial state already lies in , and infinite on a sample path that never enters . Finiteness almost surely need not imply a finite expected value. For a finite-state model, a uniform positive probability of absorption within a fixed number of steps gives a geometric series bound on the survival probabilities and proves that the mean is finite. The mean absorption time of a lazy symmetric random walk is an explicit example.
Adam bias correction 2026-11-05
Adam bias correction compensates for the initial zero values in Adam's exponentially weighted gradient histories. The basic algorithm usesThese factors normalize the coefficient sums of the two running histories.[ref]
Adam: A Method for Stochastic Optimization by Diederik P. Kingma and Jimmy Lei Ba. Algorithm1, printed p2; section3, Initialization Bias Correction, printed p3
For a scalar coordinate, suppose andExpanding the recurrence relation givesIf every has the same expectation , linearity of expectation and the geometric series yieldThus is an unbiased estimator of under these assumptions. Correlation between different times is allowed by this argument. Applying the same calculation to squared gradients with constant raw second moment gives and .
For a constant supplied gradient of 2, choose and . The first uncorrected histories are and , while their corrected values are 2 and 4. The normalization removes the missing coefficient mass caused by initialization.
It does not make a changing history equal to the latest gradient. For an original deterministic example, let , and . Then , sowhich differs from . More generally, changing moment expectations produce a weighted average of past expectations. State the moment assumptions before interpreting “bias-corrected” as unbiased for a present distribution. The nonlinear parameter update built from the corrected histories is a separate object from either moment estimator.
For real , defineThe integrand is positive and at most , so and the improper integral converges. Substituting also gives in terms of the exponential integral; the original representation makes positivity and the large- behavior transparent.
Apply the exact finite geometric series identity inside the integral, without assuming that an infinite series can be integrated term by term:The Gamma integral gives . Therefore, with retained terms,This remainder formula proves the asymptotic expansion at every fixed order: as . Its sign also gives alternating bounds: an odd overestimates , and an even underestimates it.
At , an independent evaluation of the integral gives . Three retained terms giveFive retained terms give with error bounded by . The actual errors are about and , respectively. The error bound is conservative, but established without relying on the numerical integral evaluation.
Nevertheless, the formal infinite series diverges at every fixed . The magnitudes satisfy , so they eventually increase without bound. In particular, the terms fail the necessary zero-term condition for a convergent series. The finite remainder identity remains valid after convergence has failed; this is precisely what makes the example useful.
Aggregate analysis 2026-10-18
Aggregate analysis bounds the total cost of a whole sequence directly, then divides by its length to obtain an amortized bound. Summing the copied capacities as a geometric series in dynamic array doubling is an example.
Analysis Created 2026-09-24 Updated 2026-11-05
Analysis studies limits, continuity, derivatives and integrals, and the spaces in which these operations make sense. Its central question is how information about nearby or approximating objects controls an exact answer. Real analysis makes calculus rigorous; complex analysis studies holomorphic functions; functional analysis treats entire functions as points in spaces such as a Banach space or a Hilbert space.
For example, the geometric series gives . Passing to the limit gives only for . The restriction is part of the theorem: algebra supplies the finite identity, while analysis justifies the infinite operation. This distinction matters whenever differential equations or numerical methods replace exact quantities by approximations.
Analytic continuation Created 2026-09-24 Updated 2026-10-14
An analytic continuation extends a holomorphic function beyond its original domain without changing its existing values. Formally, if are domains and is holomorphic on , a holomorphic function on with is an analytic continuation. More locally, holomorphic functions on overlapping open sets continue one another when they agree on the overlap, allowing a chain of patches along a path. The identity theorem for holomorphic functions forces agreement throughout a connected overlap once they agree on a nonempty open set in it. It also makes an extension to a fixed connected domain unique. [ref]
https://ocw.mit.edu/courses/18-04-complex-variables-with-applications-spring-2018/af8c9c78c3915aaf3fd1695f8d32a3b3_MIT18_04S18_topic13.pdf. Orloff, MIT 18.04 Topic 13, section 13.2 and Theorem 13.2
The geometric series illustrates why a formula can reach farther than the power series that first represents it. For , its partial sum isWhen , , so the series converges to . The rational function is holomorphic on and agrees there with the sum on the unit disc. It therefore continues to every point except its simple pole at . In particular,Thus analytic continuation gives a value at without making the original geometric series converge there. Its radius of convergence remains : on the terms do not even tend to zero.
The original geometric series converges only inside the unit circle, while its rational continuation exists everywhere except the pole at 1
. Details. The pole at limits the original Taylor series, although most points of its convergence circle are regular for . Recentring at gives another genuinely convergent power series,At this yields , even though the original powers of grow. This is the useful idea: overlapping local descriptions can represent one holomorphic function on a larger region.
Uniqueness on a connected overlap does not imply that every path leads to the same terminal germ of a holomorphic function. Continuing a complex logarithm around changes its value by , giving monodromy; the return patch can carry a different branch of a multivalued function. If two patches have a disconnected overlap, agreement on one component imposes no agreement on the others. [ref]
https://ocw.mit.edu/courses/18-04-complex-variables-with-applications-spring-2018/af8c9c78c3915aaf3fd1695f8d32a3b3_MIT18_04S18_topic13.pdf. Orloff, section 13.2.1, Example 13.4
Continuation using the rational function is single-valued: any path avoiding gives the same terminal germ of a holomorphic function, even if it winds around the pole. Array (data structure) 2026-10-13
A data structure storing entries at numbered positions, usually starting at index zero. The fixed-capacity representation considered here puts equal-sized slots consecutively in computer memory. The number of occupied slots can be smaller than the capacity; unused slots provide room for later entries. An array of references stores consecutive references, while the objects they identify can be elsewhere. [ref]
https://opendatastructures.org/ods-java/2_Array_Based_Lists.html. Pat Morin, Open Data Structures, chapter 2: array-based lists
Let the first slot have abstract address and each slot occupy address units. Then indexed access computesFor toy values , , the slots for start at ; index goes directly to through . These are illustrative slot addresses, not observed allocation addresses. Under the word-RAM (random-access machine) model for time complexity, word-sized arithmetic operations and any addressed computer memory access have constant cost, and the addresses and entries fit in words. Thus one array access takes time in Big O notation, independently of the index. This is the access assumption needed by binary search. [ref]
https://opendatastructures.org/ods-java/1_4_Model_Computation.html. Pat Morin, Open Data Structures, section 1.4: the word-RAM model
Inserting after in an array holding with one unused slot produces : first copy from slot to slot , then write into slot . Writing first would lose unless it had been saved separately. More generally, insert at index by copying occupied slots from right to left, then fill the gap; with spare capacity this moves entries and takes time in Big O notation, so insertion near the middle has time complexity. Removing an entry closes the gap by shifting later entries left. [ref]
https://opendatastructures.org/ods-java/2_1_ArrayStack_Fast_Stack_O.html. Pat Morin, Open Data Structures, section 2.1.1: indexed access and shifts
A fixed-capacity array cannot accept another entry when full. A resizable representation can allocate a larger array and copy the entries. Doubling capacity when full gives append cost by amortized analysis over a sequence starting empty, but a single append that copies entries still has worst-case time complexity. The outline is a geometric series: copied capacities sum to , less than twice the number of appends that triggered the last copy. Middle insertion still pays for its shifts. [ref]
https://opendatastructures.org/ods-java/2_1_ArrayStack_Fast_Stack_O.html. Pat Morin, Open Data Structures, sections 2.1.2–2.1.3: resizing and amortized costs
Asymptotic expansion Created 2026-09-24 Updated 2026-10-14
An asymptotic expansion describes a function near a specified limit by successively smaller scales. Each fixed finite truncation has a remainder smaller than its last retained scale. The central question is how the approximation improves as the parameter approaches the limit; convergence of the entire infinite series at a fixed parameter is a separate question. For the inverse-power scale ,means that, for every fixed nonnegative integer ,The little o notation makes the accuracy requirement precise. With expansions to all orders, this is also equivalent to an remainder after the th term; the implied bound may depend on .[ref]
For a quick convergent example, the finite geometric series identity givesFor , both the asymptotic expansion as and the ordinary convergent geometric series describe this function. The exact remainder displays the required extra inverse power.
The more surprising example under a divergent factorial series can approximate a convergent integral has a convergent integral but a divergent series at every fixed positive :A few terms can estimate well, even though adding sufficiently many terms eventually destroys the approximation. Optimal truncation explains where to stop.
This makes asymptotic expansions useful when an exact expression is a complicated integral or differential equation solution, but a parameter is very large or very small. The leading term reveals the scale; corrections reveal which effects matter next. The asymptotic sequence must be specified: integer powers, fractional powers, logarithms, and exponential functions can encode different kinds of behavior. A good finite approximation includes a remainder estimate or an independently checked range of use.
Autocovariance of an AR(1) process 2026-10-14
For the stationary causal AR(1) with , matching terms in its infinite moving-average representation gives, for ,Symmetry of the autocovariance extends this to negative lags. Dividing by gives the autocorrelation . The geometric series explains both the enlarged stationary variance and the diminishing memory: all past disturbances contribute to the present, but their squared weights have a finite sum. At , the process reduces to shifted white noise and all nonzero-lag autocorrelations vanish. See Bartlett, lecture 4: AR(1) representation and mean-square convergence.
Binary expansion Created 2026-09-24 Updated 2026-11-05
Every real number has a binary expansionFor , construct it by repeated doubling. Set , then and , using the floor function. Each remainder stays in , and induction givesThe remaining error is less than , so the finite partial sums converge to . For , choose every binary digit equal to one and use the geometric series.
An expansion need not be unique: . For , this ambiguity occurs precisely at the dyadic rationals, through a terminating expansion and an alternative ending in ones. At the endpoints, the displayed purely fractional form uses all zeros for zero and all ones for one. A general nonnegative real number is represented by adding the finite binary representation of its floor to an expansion of its fractional part; a negative real number is handled by a separate sign. In that broader notation, too.
Binary search tree Created 2026-10-13 Updated 2026-10-14
A binary search tree stores key-value entries in a binary tree, with distinct keys drawn from a total order. At every vertex, all keys in its left subtree must be smaller than its own key, and all keys in its right subtree must be larger. Thus the entries represent a partial function from keys to values: a key has at most one associated value and may be absent. [ref]
Lookup starts at the root of a tree. Compare the target with the current key: equality returns its value, a smaller target continues into the left subtree, and a larger target continues into the right subtree. Reaching an empty subtree reports absence. The ordering rule justifies discarding the other subtree at every step; this is the algorithm correctness argument for lookup. It applies to the entire subtree, even if that subtree contains many vertices. [ref]
In the seven-key example, search for . At , the comparison selects the right subtree; at , selects the left subtree; at , equality ends the search. The visited keys are , three key comparisons. The right-child chain stores exactly the same keys and satisfies the same binary search tree rule, but lookup of visits , five comparisons. Searching for absent key visits in the first binary search tree and all seven keys in the chain before reaching an empty right subtree.
Lookup of key in two binary search trees containing the same seven keys
. Orange highlights visited keys and edges; green marks the match. The balanced shape has height of a binary tree and requires three visits, while the chain has height of a binary tree and requires five visits for this target. Details.For a nonempty binary search tree of height of a binary tree , lookup inspects at most vertices. Under constant-cost key comparisons and child-reference access, its time complexity is in Big O notation, usually written for . With entries, , giving worst-case lookup. If the shape is kept balanced so that , lookup takes for . The empty and one-entry cases take constant work. The worst-case linear bound also appears in Sedgewick and Wayne's ordinary, unbalanced implementation. [ref]
https://raw.githubusercontent.com/kevin-wayne/algs4/master/src/main/java/edu/princeton/cs/algs4/BST.java. Sedgewick and Wayne, Algorithms: BST implementation, complexity contract and get method
In a shape with every level full, there are vertices, a geometric series. This explains how a small height of a binary tree can accommodate many entries. The binary search tree ordering rule alone does not guarantee this shape or logarithmic time complexity. Breadth-first search also gains no logarithmic key-lookup guarantee merely from having a binary tree: in the displayed balanced example, its left-to-right level order is , so it inspects six keys before finding , and an absent target requires all seven. Directed lookup uses the total order to choose one branch.
The ordering requirement includes every ancestor on the graph path from the root of a tree. Consider the smaller binary tree with root key , left child , right child , and right child under . Key must satisfy both , from its immediate parent in a rooted tree, and , from its position in the root's left subtree. Replacing by preserves all immediate parent-child comparisons, including , but violates . Lookup for would incorrectly go from to and then to an empty left subtree, overlooking the stored key.
The binary search tree ordering rule constrains an entire subtree
. The valid example has . The invalid example passes immediate parent-child comparisons, but key lies in the left subtree of key and violates its inherited upper bound. Details.An inorder traversal visits a left subtree, then its own vertex, then a right subtree. On the valid small example its output is ; on the invalid example it is , exposing the descent from to . Both seven-key binary search trees yield . The theorem is that an inorder traversal of a finite binary search tree with distinct keys lists every key once in strictly increasing total order. For a short proof outline, use mathematical induction on the number of vertices: each smaller subtree produces its own sorted list, and the full ordering rule places every left key before the current key and every right key after it. Concatenating these three parts preserves strict order. Sedgewick and Wayne's
keys method uses this left-current-right order. [ref]https://raw.githubusercontent.com/kevin-wayne/algs4/master/src/main/java/edu/princeton/cs/algs4/BST.java. Sedgewick and Wayne, Algorithms: BST keys method and inherited-bound isBST check
This combination is useful and attractive: one data structure supports directed lookup and ordered enumeration. An inorder traversal costs time complexity for entries, since it visits every vertex; balance accelerates lookup but cannot eliminate the work of outputting all keys.
// Destination: computer-science.bigb
Black-body radiation 2026-10-14
Black-body radiation is the electromagnetic radiation in thermal equilibrium with an ideal blackbody. It can be described as a photon gas: a cavity whose walls are all at the same temperature contains radiation with the universal Planck's law spectrum. A small opening in such a cavity approximates a blackbody emitter because incoming radiation undergoes repeated absorption inside. The radiation's equilibrium spectrum depends on temperature, rather than the wall material. [ref]
https://openstax.org/books/university-physics-volume-3/pages/6-1-blackbody-radiation. OpenStax, University Physics 3, §6.1, blackbody cavity and emission laws
There are three useful quantities to keep distinct: directional spectral radiance or , total energy density in the cavity, and outward radiant exitance from its opening. Planck's law supplies the first; integrals over frequency and directions supply the others. For isotropy, every direction has the same , so the full solid angle givesHere is the reduced Planck constant, and the radiation constant is . The explicit Boltzmann constant makes this an SI formula; adopting natural units with recovers .
For a short integral outline, substitute and useThe expansion is a geometric series with nonnegative terms; repeated integration by parts evaluates each integral, and is a value of the Riemann zeta function. This gives the coefficient above.
The photon gas also exerts radiation pressure, . The factor comes from the angular average of the squared directional component of photon momentum, . The thermodynamic derivation of blackbody radiation pressure gives the same result. In contrast, outgoing power involves one projected-direction factor over an outward hemisphere, yielding radiant exitanceThis is the Stefan–Boltzmann law with unit emissivity; is the Stefan-Boltzmann constant. The factor follows from Lambert's cosine law, rather than the full-sphere factor used for energy density.
For example, an ideal cavity has , radiation pressure , and outward radiant exitance . A stipulated cavity has one-sixteenth each value because all three scale as . These describe stored energy, mechanical pressure, and outgoing power per area, respectively. In a closed equilibrium cavity, equal opposing radiation balances the directional power flow even though and are nonzero.
The usefulness of black-body radiation is this common reference: Planck's law, Wien's displacement law, and the Stefan–Boltzmann law consistently connect spectral measurements, temperature, and total emission. The three illustrative curves in Planck's law are ideal equilibrium spectra; applying them to a material requires considering its emissivity and whether thermal equilibrium is a suitable approximation.
Bose-Einstein distribution Created 2026-09-24 Updated 2026-11-05
The Bose-Einstein distribution gives the mean occupation of a one-particle energy eigenstate by noninteracting bosons in thermal equilibrium:Here is the chemical potential and the temperature. It is a mean particle count in one state, so it can exceed one.
For that state, occupation numbers have weights , with . The one-state grand partition function is the geometric seriesConsequently , giving the distribution. Convergence requires for every included state, hence for a finite ideal gas with lowest energy . [ref]
At , the mean occupation is one; at much larger , it approaches , the dilute classical limit. Approaching requires careful treatment of the ground-state occupation in Bose-Einstein condensation.
Bose integral Created 2026-09-24 Updated 2026-10-14
Boundary convergence of a power series Created 2026-10-14 Updated 2026-10-18
Knowing the radius of convergence does not settle convergence on its boundary. The three power seriesall have radius one, but behave differently on .
The same radius of convergence permits three different behaviours on the unit circle
. Green denotes absolute convergence, orange conditional convergence, and red divergence. All three series converge absolutely in the open unit disk. On its boundary, the geometric series diverges everywhere; the series converges conditionally except at ; the series converges absolutely everywhere. Details.For the geometric series, boundary terms all have complex modulus one, violating the term test for divergence. For the second series, absolute values give the divergent harmonic series at every boundary point. At the series itself is harmonic and diverges. At any other point on the unit circle, its geometric partial sums obeyThe Dirichlet test with the decreasing sequence therefore gives conditional convergence. In particular, gives the negative of the alternating harmonic series, with sum . For the third series, the convergent p-series bounds the absolute values at every boundary point, giving absolute convergence by the comparison test for series.
At , the rational function is well defined and equals , yet its geometric series diverges there. A formula for a function and a particular series representation can have different domains. Likewise, applying termwise differentiation of a power series preserves the radius but need not preserve the boundary convergence.
If the terms of a power series at a nonzero point are bounded by , then . A convergent geometric series and the comparison test for series give absolute convergence at every point strictly inside that circle. Consequently terms at every point outside the radius of convergence must be unbounded. This conclusion is stronger than merely saying that those terms fail to tend to zero.
Canonical height of an elliptic curve Created 2026-09-24 Updated 2026-11-05
For an elliptic curve over a number field, fix the absolute logarithmic -height , with . The normalization used here isThe factor makes this the Néron–Tate height associated with the degree-one divisor . Some sources omit and obtain twice this function; formulas and the regulator of an elliptic curve must use a consistent convention.[ref]
Elliptic Curves by James S. Milne (2nd edition). Chapter IV, Section 4, Lemmas 4.4–4.6 and the normalization footnote after Lemma 4.6
The estimate makes successive terms differ by a summable geometric series, proving convergence and . The resulting quadratic form satisfies and the height parallelogram identity. It is nonnegative and vanishes exactly on torsion points of an elliptic curve; bounded-height rational points over a fixed number field form a finite set. For an isogeny , the corresponding normalized heights satisfy .
Cauchy identity for symmetric functions 2026-10-14
Expand each factor as a geometric series indexed by a nonnegative matrix. The RSK correspondence preserves its row and column contents, changing the sum to pairs of Semistandard Young tableaux of the same shape and proving the displayed formal identity. Its kernel is also . These are dual-basis kernels for the Hall inner product of symmetric functions, so the Schur expansion proves orthonormality. Applying the symmetric-function involution in one alphabet gives the dual identity .
Characteristic-function derivative under an absolute first moment Created 2026-10-14 Updated 2026-10-20
If , the characteristic function satisfies . The bound and the dominated convergence theorem justify differentiation at zero. For mean zero, extends continuously by , and by a finite geometric series. This supplies a direct characteristic-function proof of the weak law of large numbers.
// Target: martingale.bigb






