Algorithm 1. Sampling. number of data per class is greater than or equal to the average

• client executes: • Input ${\mathit{n}}_K$, $r \mathrm{round} \mathrm{index}$, $\delta \mathrm{oversampling} \mathrm{exponent}$ • Output ${\mathit{n}}_K$ 1: $t_k\leftarrow \frac{1}{L}{\displaystyle \sum }_{𝓁}^L{n}_k^{𝓁}\times e^{-\delta r}$ 2: repeat 3: oversampling for ${\mathcal{D}}_k$ 4: until $n_k^{𝓁} \ge$ $t_k, 𝓁\in \left[1,L\right]$ 5: return ${\mathit{n}}_K$ • server executes: • Input    $\mathcal{S}$ selected client set • Output $\delta$ 1: if $\frac{{{\displaystyle \sum }}_{k\in \mathcal{S}}^k\left({{\displaystyle \sum }}_{𝓁}^L{n}_k^{𝓁}-\left|{\mathcal{D}}_k\right|\right)}{{{\displaystyle \sum }}_{k\in \mathcal{S}}^k\left|{\mathcal{D}}_k\right|}>{\theta }_{\mathrm{over}}$ //calculate oversample data rate 2: $\delta \leftarrow \delta +\Delta$ 3: return $\delta$