SKYMIND INTELLIGENCE LAYER (SKIL)
REFERENCE ARCHITECTURE
Overview
● Why am I up here?
● Reinforcement Learning
● Use cases
● Demo!
● Deep Reinforcement Learning
● Rl4j
● Dl4j
● Spark/RL - why?
Why am I up
here?
Wrote this -->
Book Giveaway!
Reinforcement
Learning
● Learn a “policy” with repeated trial
and error
● An agent explores a search space
● Learns from rewards and penalties
each time it takes a step
● Think of win/lose scenarios
● Rewards/punishment set by an
“environment”
Credit:
http://ai.berkeley.edu/reinforcement.ht
ml
Use cases (not
games!)
● Risk analysis (loans)
● Network Intrusion
● Learning patterns from
simulations (MCMC)
Demo!
Cartpole (Hello
world of RL)
Deep
Reinforcement
Learning
● Teach a neural net from environment
● Policy determines gradient descent steps
● Most work has been based on raw frames
from games (pixel input)
● Various techniques (A3C,Policy Gradients,Deep
Q,..)
● Core idea: Neural net has a softmax
(probability distribution) mapped to actions to
take in an environment
RL4j
● Deep Reinforcement Learning
library for Java
● Openai Gym Intregration
● Deep Reinforcement Learning
with DL4j
● Implementations of A3C,DeepQ,
Policy Gradients
● Openai Gym Java Bindings
Dl4j
Dl4j
● Import keras models
● Focus on running in production
● Integrate with existing big data ecosystem
● Transparent usage of cpus and gpus
● End to end ecosystem for building data
products (not just algorithms!)
Spark/RL Why?
● Spark is distributed compute
● A lot of simulations and
environments to run
● Distributed workers running
experiments in parallel
● Data Parallelism with neural nets
Summary
● Spark for orchestrating simulations
● Spark for distributed training
● Integrated storage with HDFS
● Orchestrate GPU based spark jobs
● Easy to hook in to production (java/scala)
● Great streaming ecosystem for incremental
updates
Distributed deep rl on spark   strata singapore

Distributed deep rl on spark strata singapore

  • 2.
    SKYMIND INTELLIGENCE LAYER(SKIL) REFERENCE ARCHITECTURE
  • 3.
    Overview ● Why amI up here? ● Reinforcement Learning ● Use cases ● Demo! ● Deep Reinforcement Learning ● Rl4j ● Dl4j ● Spark/RL - why?
  • 4.
    Why am Iup here? Wrote this --> Book Giveaway!
  • 5.
    Reinforcement Learning ● Learn a“policy” with repeated trial and error ● An agent explores a search space ● Learns from rewards and penalties each time it takes a step ● Think of win/lose scenarios ● Rewards/punishment set by an “environment” Credit: http://ai.berkeley.edu/reinforcement.ht ml
  • 6.
    Use cases (not games!) ●Risk analysis (loans) ● Network Intrusion ● Learning patterns from simulations (MCMC)
  • 7.
  • 8.
    Deep Reinforcement Learning ● Teach aneural net from environment ● Policy determines gradient descent steps ● Most work has been based on raw frames from games (pixel input) ● Various techniques (A3C,Policy Gradients,Deep Q,..) ● Core idea: Neural net has a softmax (probability distribution) mapped to actions to take in an environment
  • 9.
    RL4j ● Deep ReinforcementLearning library for Java ● Openai Gym Intregration ● Deep Reinforcement Learning with DL4j ● Implementations of A3C,DeepQ, Policy Gradients ● Openai Gym Java Bindings
  • 10.
  • 11.
    Dl4j ● Import kerasmodels ● Focus on running in production ● Integrate with existing big data ecosystem ● Transparent usage of cpus and gpus ● End to end ecosystem for building data products (not just algorithms!)
  • 12.
    Spark/RL Why? ● Sparkis distributed compute ● A lot of simulations and environments to run ● Distributed workers running experiments in parallel ● Data Parallelism with neural nets
  • 13.
    Summary ● Spark fororchestrating simulations ● Spark for distributed training ● Integrated storage with HDFS ● Orchestrate GPU based spark jobs ● Easy to hook in to production (java/scala) ● Great streaming ecosystem for incremental updates