We want to learn a Baysian Network’s parameters from data.

• unbiased parameter learning
• Maximum Likelihood Parameter Learning
• Baysian Parameter Learning

If we want to do it in a Bayes Net:

parameter learning in Baysian Network

1. tag EEG by data type (what mental stage does it come from?)
2. per region, per data type, we take a band-power series
3. calculate statistics per series
4. shove the results into something interpretable

Conclusion

N1 results performs the best across brain regions; where the data came from didn’t change performance by much.

PARRY is if ELIZA had mental states such as fear, anger, and mistrust. Mentions of various things in the user turn increases or decreases each variable

Differential Equations in more than one independent variable:

\begin{equation} f(x_1, \dots, x_{n}) \end{equation}

Partially Observable Markov Decision Process is a with .

Components:

• • states
  • actions (given state)
  • transition function (given state and actions)
  • reward function
• Belief System
  • beliefs
  • observations
  • observation model \(O(o|a,s’)\)

As always we desire to find a \(\pi\) such that we can:

\begin{equation} \underset{\pi \in \Pi}{\text{maximize}}\ \mathbb{E} \qty[ \sum_{t=0}^{\infty} \gamma^{t} R(b_{t}, \pi(b_{t}))] \end{equation}

whereby our \(\pi\) instead of taking in a state for input takes in a belief (over possible states) as input.