ECS170-D10

📗 -> 06/05/25: ECS170-D10

---

Logic Review Slides
Prob and ML Slides

🎤 Vocab

❗ Unit and Larger Context

Small summary

✒️ -> Scratch Notes

Logic

Probability

Cause and Effect
Bayes Law
Naive Bayes

• Properties
• Difference between generative usages and discriminative
  MAP, MLE
• MAP - Maximum A Posterioriori Decision Rule (MAP)
• MLE - Maximum Likelihood Estimation
  • Maximizes data likelihood
    Smoothing
• Why do this?
  • Calculations don’t collapse
• Add-one Laplace smoothing
  • Completely understand how to do
• Pros and Cons
  Evaluation Metrics:
• Operator decisions
• Accuracy and…
• Recall
• Precision
• F1
  Bayesian Network
• Nodes? arrows?
• Full specification (syntax)

Cause to effect, effect to cause
Complexity of Exact Inference Networks

Markovs

Markov Property

• Current state only dependent on immediately preceding state
  Higher order markovian chain
• Dependent on previous TWO states
  • Second-order markov chain in this example
    Formal specification
• Parameters, what different matrices stand for ()

Hidden Markov Models

Algorithms

Ooh boyy…

• Forward algorithm
• Decoding
• Viterbi
• Viterbri vs. Forward
• Learning
• Forward-backward
• Language modeling
• Perplexity
  • What does it mean when it is low and high?

For the ml sections, it is almost perfectly summarizing what the final will test of ML

ML

Type of ML
Supervised/unsupervised
Input of each algo
Overfitting
Underfitting
Bias, variance

• What does this mean?
  Complex models
• How does this affect bias/variance?
  Effect of more layer on bias
  Effect of regularization on bias
  Different dataset: training, validation, test

---

Questions

Is minimax on final?

yes

constrain satisfaction on the final?

yes (fast)

monte carlo search?

yes (instant)

space and time time complexity of algos?

yes

variable elimination algorithm?

for bayesian, no
(did not specify for logic)

questions about bayesian networks, but not the calculations
more conceptual

specific proof resolution algorithms?

maybe??

A* algo?

yes (a lil slower)

hill climbing?

yes

does the minimax involve pruning?

maybe

non deterministic environments? (i think?)

no

a conceptual question about markov model

• know the properties

understand notation of complexity

estimating parameters?

yes

🧪 -> Refresh the Info

Did you generally find the overall content understandable or compelling or relevant or not, and why, or which aspects of the reading were most novel or challenging for you and which aspects were most familiar or straightforward?)

Did a specific aspect of the reading raise questions for you or relate to other ideas and findings you’ve encountered, or are there other related issues you wish had been covered?)

🔗 -> Links

Resources

• Put useful links here

Connections

• Link all related words