What do you mean by sequential prediction in your paper?

[Muhammad Sarim Mehdi]

Hello. I read in your paper that you did an ablation study where you removed the sequential prediction and replaced it with parallel prediction where you predicted all three time steps with a different fully connected layer. I understand parallel prediction but I don't understand what you really mean by sequential prediction.

By sequential prediction do you mean that at every time step you use the same fully connected layer to do the prediction or that you use one fully connected layer at the final time step only?

[Weidi]

Hi,

I'll use the Figure 2 in the paper to explain this.

As you mentioned, parallel prediction means to predict ẑ_t+1, ẑ_t+2, ..., with separate FC layers.

While for sequential prediction, it's simply meaning an autoregressive model, so we use the convGRU. In detail, the context vector c_t computed from z_t and hidden states from previous step (not shown in the figure), is used to predict the ẑ_t+1.

Then the predicted ẑ_t+1, and hidden states are further applied to compute c_t+1,. Ideally, now c_t+1, will include all the information from t+1 step, and it can be used to predict ẑ_t+2,