Lecture 2

• Lecture 2
  • Video
  • Can we build hardware as complex as the brain
  • Must an intelligent System be Foolproof
  • Some questions
  • Approaches to AI
  • Planning
  • Exploration vs Exploitation
  • Action selection strategies
  • What is search
  • Problems solving through search

Video

link

Can we build hardware as complex as the brain

• We would like AI to have brain of human as well as animal
  • ant colony optimization
    • how ants locate a food source and how they communicate it to fellow ants in neighbor
  • swarm intelligence
    • swarm = collection of animals/birds/fishes
    • how they communicate and do a task is swarm intelligence
• How complicated is our brain?
  • a neuron is basic info processing unit
  • ~10^12 neurons in a human brain
  • many more synapses (10^14) connecting these neurons
  • cycle time = 10^-3 seconds (1ms)
• How complex we can make computers
  • 10^8 or more transistors per CPU
  • supercomputer hundreds of CPUs, 10^9 bits of RAM
  • cycle times = order of 10^-8 seconds
• Conclusion

Must an intelligent System be Foolproof

Some questions

Approaches to AI

Planning

• planning a policy = considering the future consequences of actions to choose the best one

• At evry step, I will try to take greedy approach(pick the best option available)
  • greedy approach will give me suboptimal solutions in most cases
  • why we approach greedy?
    • bcz search space is so large, that we cannot enumerate everything
    • so I decide a heuristic

Exploration vs Exploitation

• exploitation - exploiting ur knowledge to make the best decision
• do not worry about short term gains but overall
• Real life example

Action selection strategies

• dotted lines are the best options
• greedy me I will use this only
  • prob of picking this in greedy = 1.0
  • baki options ki prob = 0
• eps-greedy
  • eps = 0.1 => 90% of time I will take best option
  • and with remaining prob, I will take remaining options
• Soft-max

What is search

Problems solving through search