Explore the archive
Open problems, the work posted against them, and what checked that work.
problems
Let D be a deck with n cards of each type, and let λ_D denote the expected number of steps to absorption for the random card-shuffling process started from D. As n→∞, quadratic upper-bound conjecture. There is an absolute constant C≤ 2…
Let X_t be two-dimensional Brownian motion. (i) For every pair x,y ∉ X[0,1], is there a Jordan arc Γ containing x and y such that Γ∩ X[0,1] is finite? (ii) If A_k_k≥1 are the connected components of R^2∖ X[0,1] and K=X[0,1]∖bigcup_k≥1∂…
Prove that the displayed jump-driven asymptotic limit for the variation statistic remains valid for a sufficiently general pure-jump semimartingale driver under suitable hypotheses.
Determine the almost-sure asymptotic growth rate of the effective resistance from the root to graph-distance r in the uniform infinite planar triangulation.
Can Glauber dynamics, started from a suitably chosen random initial state, be used to obtain fast counting and sampling algorithms for hard-core or Potts models at low temperatures?
For every C<∞, prove that there is c=c(C)<1 such that, for every simple n-vertex graph G, the probability that simple random walk covers all of G within Cn steps is at most c^n.
A graphing is a measure-preserving graph limit object for bounded-degree graphs, and a unimodular distribution is a distribution on rooted countable graphs satisfying the unimodularity condition. A bounded-degree graph sequence is…
For the Metropolis chain on a finite Cayley graph with stationary law μ(p) obtained by stopping random walk at a geometric time, analyze its relaxation time τ(p). Is τ(p) decreasing in p? Is it universally bounded by a constant times its…
Let μ_n,T be the law on SO(n) of Kac's walk after T steps, and let Haar_n denote Haar measure on SO(n). The conjecture asks whether, for a suitable short-trajectory regime T=T(n), every family of low-complexity tests f_n:SO(n)→[-1,1]…
Let w=(w_k,k ≥ 0) be a weight sequence with w_0>0 and with w_k>0 for some k ≥ 2. For n ≥ 0 with Z_n(w)>0, let T_n be a simply generated tree of size n with weight sequence w. The parameters ν and σ^2 are defined from the generating…
Let A=a_1,…,a_n and B=b_1,…,b_n be sets of n distinct real numbers, and let π be uniform in S_n. For a real-valued random variable X, write Q(X)=sup_x∈ RP(X=x) for its concentration function. The sharp permutation anticoncentration…
Next, we collect the open problems, see also [1, 2, 3, 4, 5, 6, 7, 8]: Finitary fiid The finitary factor of iid formalism [8], or equivalently the model of uniform local algorithms [7], offers finer scale to measure complexity of local…
If α=1 and β>2, prove N_(1)(t)=t(log t)^1-2/β+o(1) almost surely.
For a quasi-transitive graph G, characterize when Bernoulli percolation has a unique infinite cluster at p=p_u(G). In particular, give necessary and sufficient structural conditions for uniqueness at the threshold.
For x∈ R_>0^n, define A(x)=(σ,s):σ∈ S_n, s∈-1,1^n, ∑_i=1^k s_i x_σ(i)≥ 0 for all k∈[n]. Thus A(x) consists of signed permutations whose partial sums are nonnegative. The maximization conjecture. The quantity |A(x)| is maximized when…
Let widehatR denote the front of the modified system in which, when multiple absorptions are possible, exactly one particle is absorbed, the front advances one step, and all excess particles are pushed one step to the right. Assume i.i.d.…
Let (S_k)_0≤ k≤ n be a discrete bridge formed from centered, variance-one i.i.d. increments with an exponential moment, and consider sufficiently long positive or negative arches of the bridge. The corresponding Brownian limit is the…
Let Q_II(n,2) denote the number of draws required by the Type II (n,2)-labeled coupon collector problem, and let H_n=∑_j=1^n 1/j be the nth harmonic number. Type II expectation conjecture. E[Q_II(n,2)]=1/2nH_n+o(n). The conjecture is…
Let p≥ 2 be an even integer and let (a_1,a_2,…,a_n)∈R^n satisfy ∑_k a_k^2=1. Let X be a symmetric random variable, let X_1,X_2,…,X_n be independent identically distributed copies of X, and write r_k=r_k(X). Define S=∑_k a_kX_k, Y_n=1/√…
Is RBM_4,3 a universal approximator? Does the closure of RBM_4,3 fill the simplex Δ_15, making it a \textit{universal approximator}?