#line 1 "modint_static.test.cpp"
#define PROBLEM "https://yukicoder.me/problems/no/2062"

// 提出時にassertはオフ
#ifndef DEBUG
#ifndef NDEBUG
#define NDEBUG
#endif
#endif

#include <bits/stdc++.h>
#line 2 "/home/cocojapanpan/Procon_CPP/proconLibrary/myLibrary/modint_static.hpp"

#line 2 "/home/cocojapanpan/Procon_CPP/proconLibrary/myLibrary/innermath_modint.hpp"

#line 4 "/home/cocojapanpan/Procon_CPP/proconLibrary/myLibrary/innermath_modint.hpp"

#ifdef _MSC_VER
#include <intrin.h>
#endif

namespace innermath_modint{
    using ll = long long;
    using ull = unsigned long long;
    using u128 = __uint128_t;

    // xのmodを[0, mod)で返す
    constexpr ll safe_mod(ll x, ll mod) {
        x %= mod;
        if (x < 0) x += mod;
        return x;
    }

    constexpr ll pow_mod_constexpr(ll x, ll n, ll mod) {
        if (mod == 1) return 0;
        ll ret = 1;
        ll beki = safe_mod(x, mod);
        while (n) {
            // LSBから順に見る
            if (n & 1) {
                ret = (ret * beki) % mod;
            }
            beki = (beki * beki) % mod;
            n >>= 1;
        }
        return ret;
    }

    // int型(2^32以下)の高速な素数判定
    constexpr bool is_prime_constexpr(int n) {
        if (n <= 1) return false;
        if (n == 2 || n == 7 || n == 61) return true;
        if (n % 2 == 0) return false;
        // ミラーラビン判定 int型ならa={2,7,61}で十分
        constexpr ll bases[] = {2, 7, 61};
        // n-1 = 2^r * d
        ll d = n - 1;
        while (d % 2 == 0) d >>= 1;
        // 素数modは1の平方根として非自明な解を持たない
        // つまり非自明な解がある→合成数
        for (ll a : bases) {
            ll t = d;
            ll y = pow_mod_constexpr(a, t, n);
            // yが1またはn-1になれば抜ける
            while (t != n - 1 && y != 1 && y != n - 1) {
                y = (y * y) % n;
                t <<= 1;
            }
            // 1の平方根として1と-1以外の解(非自明な解)が存在
            if (y != n - 1 && t % 2 == 0) {
                return false;
            }
        }
        return true;
    }

    // 拡張ユークリッドの互除法 g = gcd(a,b)と、ax = g (mod b)なる0 <= x <
    // b/gのペアを返す
    constexpr std::pair<ll, ll> inv_gcd(ll a, ll b) {
        a = safe_mod(a, b);
        // aがbの倍数
        if (a == 0) return {b, 0};
        // 以下 0 <= a < b
        // [1] s - m0 * a = 0 (mod b)
        // [2] t - m1 * a = 0 (mod b)
        // [3] s * |m1| + t * |m0| <= b
        ll s = b, t = a;
        ll m0 = 0, m1 = 1;
        while (t) {
            // s → s mod t
            // m0 → m0 - m1 * (s / t)
            ll u = s / t;
            s -= t * u;
            m0 -= m1 * u;
            std::swap(s, t);
            std::swap(m0, m1);
        }
        // s = gcd(a, b)
        // 終了の直前のステップにおいて
        // [1] k * s - m0 * a = 0 (mod b)
        // [2] s - m1 * a = 0 (mod b)
        // [3] (k * s) * |m1| + s * |m0| <= b
        // |m0| < b / s
        if (m0 < 0) m0 += b / s;
        return {s, m0};
    }

    // barret reduction 掛け算のmodの定数倍高速化(modが定数でない場合に使用)
    struct barretReduction {
       public:
        explicit barretReduction(uint _mod)
            : mod(_mod),
              imod((ull)(-1) / mod + 1) {}  // unsignedの場合、負で巡回する

        uint get_mod() const { return mod; }

        uint mul(int a, int b) const {
            ull z = a;
            z *= b;
#ifdef _MSC_VER
            ull x;
            _umul128(z, imod, &x)
#else
            // x = z / mod またはその +1
            // 割り算をビットシフトにすることで高速化
            ull x = (ull)(((u128)z * imod) >> 64);
#endif
            // z - x * mod = z % mod - mod の場合、uintなので 2^32 - (mod -
            // z % mod) つまりmodを足せば 2^32 + z %
            // modとなり、求めるmodになる
            uint v = (uint)(z - x * mod);
            if (v >= mod) v += mod;
            return v;
        }

       private:
        uint mod;
        ull imod;  // ceil(2^64 / mod)
    };
}
#line 5 "/home/cocojapanpan/Procon_CPP/proconLibrary/myLibrary/modint_static.hpp"

template <const int MOD>
struct modint_static {
    using ll = long long;

   public:
    constexpr modint_static(ll x = 0) noexcept : value(x % MOD) {
        if (value < 0) value += MOD;
    }
    constexpr int get_mod() const noexcept { return MOD; }
    constexpr ll val() const noexcept { return value; }
    constexpr modint_static operator-() const noexcept {
        return modint_static(-value);
    }
    constexpr modint_static& operator++() noexcept {
        ++value;
        if(value == MOD) value = 0;
        return *this;
    }
    constexpr modint_static& operator--() noexcept {
        if(value == 0) value = MOD;
        --value;
        return *this;
    }
    constexpr modint_static operator++(int) noexcept {
        modint_static cpy(*this);
        ++(*this);
        return cpy;
    }
    constexpr modint_static operator--(int) noexcept {
        modint_static cpy(*this);
        --(*this);
        return cpy;
    }
    constexpr modint_static& operator+=(const modint_static& rhs) noexcept {
        value += rhs.value;
        if (value >= MOD) value -= MOD;
        return *this;
    }
    constexpr modint_static& operator-=(const modint_static& rhs) noexcept {
        value += (MOD - rhs.value);
        if (value >= MOD) value -= MOD;
        return *this;
    }
    constexpr modint_static& operator*=(const modint_static& rhs) noexcept {
        (value *= rhs.value) %= MOD; // 定数だとコンパイラ最適化がかかる
        return *this;
    }
    constexpr modint_static operator+(const modint_static& rhs) const noexcept {
        modint_static cpy(*this);
        return cpy += rhs;
    }
    constexpr modint_static operator-(const modint_static& rhs) const noexcept {
        modint_static cpy(*this);
        return cpy -= rhs;
    }
    constexpr modint_static operator*(const modint_static& rhs) const noexcept {
        modint_static cpy(*this);
        return cpy *= rhs;
    }
    constexpr modint_static pow(ll beki) const noexcept {
        modint_static curbekimod(*this);
        modint_static ret(1);
        while (beki > 0) {
            if (beki & 1) ret *= curbekimod;
            curbekimod *= curbekimod;
            beki >>= 1;
        }
        return ret;
    }

    // valueの逆元を求める
    constexpr modint_static inv() const noexcept {
        // 拡張ユークリッドの互除法
        auto [gcd_value_mod, inv_value] = innermath_modint::inv_gcd(value, MOD);
        assert(gcd_value_mod == 1);
        return modint_static(inv_value);
    }
    constexpr modint_static& operator/=(const modint_static& rhs) noexcept {
        return (*this) *= rhs.inv();
    }
    constexpr modint_static operator/(const modint_static& rhs) const noexcept {
        modint_static cpy(*this);
        return cpy /= rhs;
    }

   private:
    ll value;
};

using mint998244353 = modint_static<998244353>;
using mint1000000007 = modint_static<1000000007>;
#line 12 "modint_static.test.cpp"
using mint = mint998244353;
constexpr int subset_mod = 999630629;
using namespace std;
using ll = long long;

#define ALL(x) (x).begin(), (x).end()
template <class T> using vec = vector<T>;

int main() {
    ios_base::sync_with_stdio(false);
    cin.tie(nullptr);
    int N;
    cin >> N;
    vec<int> A(N);
    int all_sum = 0;
    for(int &a : A) {
        cin >> a;
        all_sum += a;
    }
    // 基本的に総和の2^(N-1)倍で終わりだが、999630629を超えると引かないとダメ
    // いくつのsubsetでこれが起きるか数えます
    ll haveToSubtract = 0;
    if(all_sum > subset_mod){
        multiset<int> sum_set;
        int last_sum = all_sum - subset_mod;
        // 要素の総和がlast_sum以下となるsubsetの個数を数える
        for(int i = 0; i < N; i++){
            multiset<int> cur_sum_set;
            for(int sum : sum_set){
                if(sum + A[i] <= last_sum){
                    cur_sum_set.insert(sum + A[i]);
                } else {
                    break;
                }
            }
            for(int cur_sum : cur_sum_set){
                sum_set.insert(cur_sum);
            }
        }
        haveToSubtract = sum_set.size();
    }
    mint ans(all_sum);
    ans *= (1 << (N - 1));
    ans -= mint(haveToSubtract) * subset_mod;
    cout << ans.val() << "\n";
}