Byszewski–Ulas digit-sum factorization conjecture

Let r1r\ge 1, let N1,,NrN_1,\ldots,N_r be positive integers, and let x1,,xr,y1,,yrx_1,\ldots,x_r,y_1,\ldots,y_r be variables. For each nonnegative integer nn, write s2(n)s_2(n) for the sum of the binary digits of nn. Byszewski–Ulas conjecture. One has

n1=02N11nr=02Nr1(1)j=1rs2(nj)(j=1rs2(nj)xj+njyj)j=1rNj=(1)j=1rNr(j=1rNj)!j=1rij=0Nj1(xj+2ijyj).\sum_{n_1=0}^{2^{N_1}-1}\cdots\sum_{n_r=0}^{2^{N_r}-1}(-1)^{\sum_{j=1}^r s_2(n_j)}\left(\sum_{j=1}^r s_2(n_j)x_j+n_jy_j\right)^{\sum_{j=1}^r N_j}=(-1)^{\sum_{j=1}^r N_r}\left(\sum_{j=1}^r N_j\right)!\prod_{j=1}^r\prod_{i_j=0}^{N_j-1}(x_j+2^{i_j}y_j).

The paper presents this as another conjecture of Byszewski and Ulas and states in the introduction that it proves and generalizes the two displayed conjectures. The identity gives a product factorization for a digit-sum-weighted power sum.

Sources & referencesView supporting material

Primary source

Tanay Wakhare and Christophe Vignat, “Settling some sum suppositions”, arXiv:1805.10569 (2018).

Progress summary

Never refreshed

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

Solutions 0

No solutions have been posted yet.