Phil 12.22.15

VTX 7:00 – 6:00

  • Probabilistic Inference II
    • Assertion – Any variable in a graph is said by me to be independent of any other non-descendant, given its parents. All the causality flows through the parents.
    • A belief net or Bays net is *always* acyclic and directed.
    • Traverse the graph from the bottom up, so that no node depends on a node to its left in a list.
    • Generating the list:BayesNetFromData
    • When using the list, work from the top down in the list
    • Naive Bayesian inference
      • P(a|b)P(b) = P(a,b) = P(b|a)P(a)
      • P(a|b) = (P(b|a)P(a))/P(b) BayesChain
      • Can use Bayes to decide between models – Naive Bayesian Classification
      • Use the sum of the logs of the probabilities rather than the products because otherwise we run out of bits of precision
    • The right thing to do when you don’t know anything (just have symptoms)
  • Hibernate
    • Adding config.setProperty(“”, “update”); to the setup, so that tables can be rebuilt on demand. Nope, that didn’t work. Maybe I can’t split configuration between the config file and programmatic variables?
    • The only way that I was able to get this to work as an argument was to have a setupTables flag indicate which config to read. That works well though.
    • Got simple collections running, which means that I should be able to get networks built. Basically modified the example from Just Hibernate that starts on page 53.
    • Next, we work on getting inheritance to work. I think this will help.
  • Initial Java class network thoughts, just to try storing and retrieving items
    • BaseItem
      • guid
    •  BaseNode extends BaseItem
      • node_id
      • name
    • BaseEdge extends BaseItem
      • edge_id
      • source
      • target
      • weight
    • BaseNetwork extends BaseItem
      • network_id
      • name
      • owner
      • edgeList
      • nodeList (we need this because we may have orphans in the network)
    • BaseOwner extends BaseItem
      • owner_id
      • name
      • password?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.