/- Copyright 2025 The Formal Conjectures Authors. Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at https://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. -/ module public import FormalConjecturesUtil

Erdős Problem 321

Reference: erdosproblems.com/321

@[expose] public sectionopen Filter Realopen scoped Finsetnamespace Erdos321

Let $R(N)$ be the size of the largest $A\subseteq{1, ..., N}$ such that all sums $\sum_{n\in S} \frac{1}{n}$ are distinct for $S\subseteq A$.

noncomputable def R (N : ) : := sSup { #A | (A) (_ : A Finset.Icc 1 N) (_ : Set.InjOn (fun (S : Finset ) n S, (1 : ) / n) A.powerset) }

Let $R(N)$ be the size of the largest $A\subseteq{1, ..., N}$ such that all sums $\sum_{n\in S} \frac{1}{n}$ are distinct for $S\subseteq A$. What is $R(N)$?

@[category research open, AMS 11] theorem erdos_321 (N : ) : R N = answer(sorry) := N:R N = sorry All goals completed! 🐙

Let $R(N)$ be the size of the largest $A\subseteq{1, ..., N}$ such that all sums $\sum_{n\in S} \frac{1}{n}$ are distinct for $S\subseteq A$. What is $\Theta(R(N))$?

@[category research open, AMS 11] theorem erdos_321.variants.isTheta : (fun N (R N : )) =Θ[atTop] (answer(sorry) : ) := (fun N (R N)) =Θ[atTop] sorry All goals completed! 🐙

Let $R(N)$ be the size of the largest $A\subseteq{1, ..., N}$ such that all sums $\sum_{n\in S} \frac{1}{n}$ are distinct for $S\subseteq A$. Find the simplest $g(N)$ such that $R(N) = O(g(N))$.

@[category research open, AMS 11] theorem erdos_321.variants.isBigO : (fun N (R N : )) =O[atTop] (answer(sorry) : ) := (fun N (R N)) =O[atTop] sorry All goals completed! 🐙

Let $R(N)$ be the size of the largest $A\subseteq{1, ..., N}$ such that all sums $\sum_{n\in S} \frac{1}{n}$ are distinct for $S\subseteq A$. Find the simplest $g(N)$ such that $R(N) = o(g(N))$.

@[category research open, AMS 11] theorem erdos_321.variants.isLittleO : (fun N (R N : )) =o[atTop] (answer(sorry) : ) := (fun N (R N)) =o[atTop] sorry All goals completed! 🐙

Let $R(N)$ be the maximal such size. Results of Bleicher and Erdős from [BlEr75] and [BlEr76b] imply that $$ \frac{N}{\log N} \prod_{i=3}^{k} \log_i N \le R(N), $$ valid for any $k \ge 4$ with $\log_k N \ge k$ and any $r \ge 1$ with $\log_{2r} N \ge 1$. (In these bounds $\log_i n$ denotes the $i$-fold iterated logarithm.)

[BlEr75] Bleicher, M. N. and Erdős, P., The number of distinct subsums of $\sum \sb{1}\spN,1/i$. Math. Comp. (1975), 29-42. [BlEr76b] Bleicher, Michael N. and Erdős, Paul, Denominators of Egyptian fractions. II. Illinois J. Math. (1976), 598-613.

@[category research solved, AMS 11] theorem erdos_321.variants.lower (N k : ) (hk : 4 k) (hkN : k log^[k] N) : N / log N * i Finset.Icc 3 k, (log^[i] N) R N := N:k:hk:4 khkN:k log^[k] NN / log N * i Finset.Icc 3 k, log^[i] N (R N) All goals completed! 🐙

Let $R(N)$ be the maximal such size. Results of Bleicher and Erdős from [BlEr75] and [BlEr76b] imply that $$ R(N) \le \frac{1}{\log 2} \log_r N \left( \frac{N}{\log N} \prod_{i=3}^{r} \log_i N \right), $$ valid for any $k \ge 4$ with $\log_k N \ge k$ and any $r \ge 1$ with $\log_{2r} N \ge 1$. (In these bounds $\log_i n$ denotes the $i$-fold iterated logarithm.)

Since Real.log is defined on all of $\mathbb{R}$, the condition $\log_{2r} N \ge 1$ is formalised as $\log_i N \ge 1$ for all $i \le 2r$, which is equivalent to it for the genuine iterated logarithm and ensures that every intermediate iterate is positive.

[BlEr75] Bleicher, M. N. and Erdős, P., The number of distinct subsums of $\sum \sb{1}\spN,1/i$. Math. Comp. (1975), 29-42. [BlEr76b] Bleicher, Michael N. and Erdős, Paul, Denominators of Egyptian fractions. II. Illinois J. Math. (1976), 598-613.

@[category research solved, AMS 11] theorem erdos_321.variants.upper (N r : ) (hr : 1 r) (hrN : i 2 * r, 1 log^[i] N) : R N 1 / log 2 * log^[r] N * N / log N * i Finset.Icc 3 r, (log^[i] N) := N:r:hr:1 rhrN: i 2 * r, 1 log^[i] N(R N) 1 / log 2 * log^[r] N * N / log N * i Finset.Icc 3 r, log^[i] N All goals completed! 🐙end Erdos321