Alphabeta Math
AlgorithmAlgorithm: Literature-sourcedProof: Not applicablePipeline-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Bias-corrected Adam update

Definition

Fix a dimension d at least 1, an initial point x_1 in R^d, a constant step size eta>0, a denominator constant epsilon>0, and parameters beta_1,beta_2 in [0,1). Set m_0=v_0=0. Given a gradient g_t in R^d at each integer time t>=1, calculate

mt=β1mt−1+(1−β1)gt,vt=β2vt−1+(1−β2)(gt⊙gt),

where \odot denotes coordinatewise multiplication. Define the bias-corrected moments

m^t=mt1−β1t,v^t=vt1−β2t,

and update, coordinatewise,

xt+1=xt−ηm^tv^t+ϵ.

All operations involving vectors in the update are coordinatewise except the ordinary vector addition in the moment recurrences. Since 1-\beta_j^t>0 for t>=1 and v_t has nonnegative coordinates, every denominator is positive; the procedure is defined at every finite step for every finite gradient sequence.

This item defines the update only. It makes no convergence or regret claim.

Remarks

The real field used here is complete by The Cauchy-sequence reals have the least-upper-bound property, so the complete-ordered-field hypothesis of Square roots exist: a unique a≥0 with (a)2=a; the positives are {x2:x≠0} holds.

For 0≤βj<1 and integer t≥1, Monotonicity of x↦xn and of n↦an gives 0≤βjt<1, so each bias-correction denominator is positive. For each coordinate, a squared gradient is nonnegative by Squares of nonzero elements are positive (with the zero case equal to zero); induction in the recurrence, using the nonnegative coefficients β2 and 1−β2 and the product sign rules Sign rules for products and monotonicity of multiplication, Multiplication by zero: 0⋅a=0, gives vt≥0. Division by the positive bias-correction factor preserves nonnegativity by Inverses of positives are positive, and reciprocation reverses order, so v^t≥0; its square root exists by Square roots exist: a unique a≥0 with (a)2=a; the positives are {x2:x≠0}, and v^t+ϵ>0. These facts justify every finite update.

Depends on

Used by

Dependency tree · two levels

37 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources