Zagier–Hoffman–Saha dimension and basis conjectures for multiple zeta values and multiple t-values

From papers

Let (dn)n1(d_n)_{n\geqslant1} be the Padovan sequence defined by d1=0d_1=0, d2=1d_2=1, d3=1d_3=1 and dn=dn2+dn3d_n=d_{n-2}+d_{n-3} for n4n\geqslant4, and let (fn)n1(f_n)_{n\geqslant1} be the Fibonacci sequence with f1=f2=1f_1=f_2=1 and fn=fn1+fn2f_n=f_{n-1}+f_{n-2} for n3n\geqslant3. Let Zn\mathcal{Z}_n and Tn\mathcal{T}_n denote the Q\mathbf{Q}-spans of all multiple zeta values and multiple tt-values of weight nn, respectively. A multiple zeta value of index (i1,,ik)(i_1,\ldots,i_k) has weight i1++iki_1+\cdots+i_k, and a multiple tt-value is denoted by t(i1,,ik)t(i_1,\ldots,i_k). Zagier–Hoffman–Saha conjectures. The following statements hold: dimZn=dn\dim \mathcal{Z}_n=d_n; dimTn=fn\dim \mathcal{T}_n=f_n; the set

{ζ(i1,i2,,ik)i1+i2++ik=n, ij{2,3}}\{\zeta(i_1, i_2, \ldots, i_k)\mid i_1+i_2+\cdots+i_k=n,\ i_j\in\{2,3\}\}

is a Q\mathbf{Q}-basis of Zn\mathcal{Z}_n; and, for n2n\geqslant2, the set

{t(a1+1,a2,,ar)a1+a2++ar=n1, ai{1,2}, 1rn1}\{t(a_1+1,a_2,\ldots,a_r)\mid a_1+a_2+\cdots+a_r=n-1,\ a_i\in\{1,2\},\ 1\leqslant r\leqslant n-1\}

is a Q\mathbf{Q}-basis of Tn\mathcal{T}_n. These conjectures concern the expected dimensions and explicit bases of the spaces of multiple zeta values and multiple tt-values; the source presents them as conjectures attributed to Zagier, Hoffman, and Saha, with no resolution supplied.

Progress summary

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Sources & referencesView supporting material

Primary source

Sarth Chavan, Masato Kobayashi and Jorge Layja, “Integral Evaluation of Odd Euler Sums, Multiple t-Value t(3,2,,2) and Multiple Zeta Value ζ(3,2,,2)”, arXiv:2111.07097 (2021).

Solutions 0

No solutions have been posted yet.