deletions | additions       diff --git a/Waveform optimization.tex b/Waveform optimization.tex  index aefad55..46b56bb 100644  --- a/Waveform optimization.tex  +++ b/Waveform optimization.tex 
...
We can now expand the integral over the hidden states:  $$\idotsint $\idotsint  \prod_{j=1}^N \prod_{u=1}^{T_j} dx_u^j \sum_i [ \sum_t (d_t^i - m(x_t^i))^2 ]P(X|D,\Theta_{k-1})$$ ]P(X|D,\Theta_{k-1})$  $$ $  = \sum_{i,t} \idotsint \prod_j \prod_u dx_u^j (d_t^i - m(x_t^i))^2 P(X|D,\Theta_{k-1})$$ P(X|D,\Theta_{k-1})$  $$ $  = \sum_{i,t} \iint dx_t^i (d_t^i - m(x_t^i))^2 \idotsint \prod_{j\neq i} \prod_{u\neq t} dx_u^j P(X|D,\Theta_{k-1})$$ P(X|D,\Theta_{k-1})$  $$ $  = \sum_{i,t} \iint dx_t^i (d_t^i - m(x_t^i))^2 P(x_t^i|D,\Theta)$$ P(x_t^i|D,\Theta)$  Where Here  $P(x_t^i|D,\Theta)$ is the posterior distribution of our hidden states at time $t$ for a trace $i$, and is computed as a discrete table indexed by integer $l \in [1,..N_\lambda]$ and $m \in [1,..N_\phi]$: $P_t^i[l,m]$ . Since we work in discretized space we rewrite the above integral as: $ \sum_i \sum_t \sum_l \sum_m (d_t^i - \exp(\lambda[l])w(\phi[m]))^2 P_t^i[l,m]$