Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 8, 2018

• Configure optimistic Q-learning to be comparable to MC-Dirichlet

  aslanides

  committed
  b4d7f9b

  Copy full SHA for b4d7f9b

• Replace 'cycles' with 'steps'

  aslanides

  committed
  a3dda3e

  Copy full SHA for a3dda3e

Commits on Jul 3, 2017

• fix hyperlink

  aslanides

  committed
  02cf492

  Copy full SHA for 02cf492

• run linter on HTML

  aslanides

  committed
  47e3e4c

  Copy full SHA for 47e3e4c

• fix beta -> this.beta in SoftQLearn

  aslanides

  committed
  694bd9c

  Copy full SHA for 694bd9c

• Merge pull request #19 from tom4everitt/master

  Show description for f30ceea
  aslanides

  authored
  f30ceea

  Copy full SHA for f30ceea

• improve Reward Corruption description

  tom4everitt

  committed
  23e2c4e

  Copy full SHA for 23e2c4e

Commits on Jun 29, 2017

• fix default Q-value initialization

  aslanides

  committed
  a26eb05

  Copy full SHA for a26eb05

Commits on Jun 28, 2017

• Merge pull request #18 from aslanides/rc1

  Show description for 6afc7b0
  aslanides

  authored
  6afc7b0

  Copy full SHA for 6afc7b0

• tidy up vis/novis logic

  aslanides

  committed
  334920b

  Copy full SHA for 334920b

• run linter

  aslanides

  committed
  1fe07ed

  Copy full SHA for 1fe07ed

• add ESLint + npm

  aslanides

  committed
  d3f39a7

  Copy full SHA for d3f39a7

• updated description for reward corruption demo

  

  vkrakovna

  authored and

  aslanides

  committed
  1724c65

  Copy full SHA for 1724c65

• added documentation to the plotting notebook

  

  vkrakovna

  authored and

  aslanides

  committed
  e498018

  Copy full SHA for e498018

• added option for plotting after running, rather than during

  

  vkrakovna

  authored and

  aslanides

  committed
  24f8b4a

  Copy full SHA for 24f8b4a

• fixed typo

  

  vkrakovna

  authored and

  aslanides

  committed
  47fa549

  Copy full SHA for 47fa549

• made download of result files optional

  

  vkrakovna

  authored and

  aslanides

  committed
  7aa3aad

  Copy full SHA for 7aa3aad

• updated default tabular agent parameters

  

  vkrakovna

  authored and

  aslanides

  committed
  cd56ba6

  Copy full SHA for cd56ba6

Commits on Jun 27, 2017

• add simple url queries for fast demo selection

  aslanides

  committed
  37789b4

  Copy full SHA for 37789b4

Commits on Jun 25, 2017

• Merge pull request #17 from aslanides/rc

  Show description for 0b64c9a
  aslanides

  authored
  0b64c9a

  Copy full SHA for 0b64c9a

Commits on Jun 22, 2017

• add some finishing touches to @vkrakovna's patch.

  Show description for a3f44d2
  aslanides

  committed
  a3f44d2

  Copy full SHA for a3f44d2

Commits on Jun 21, 2017

• created a new config for console experiments, updated description, cleaned up logging

  

  vkrakovna

  authored and

  aslanides

  committed
  cf94f6d

  Copy full SHA for cf94f6d

• added explanation for reward corruption setup

  

  vkrakovna

  authored and

  aslanides

  committed
  973d90c

  Copy full SHA for 973d90c

• added plotting and summary code for reward corruption experiments

  

  vkrakovna

  authored and

  aslanides

  committed
  5122ad0

  Copy full SHA for 5122ad0

• added quantilising and softmax agents

  

  vkrakovna

  authored and

  aslanides

  committed
  b55b253

  Copy full SHA for b55b253

• added outputs for corrupt and true rewards

  

  vkrakovna

  authored and

  aslanides

  committed
  a7a1e79

  Copy full SHA for a7a1e79

• added trace plots for true and corrupt rewards

  

  vkrakovna

  authored and

  aslanides

  committed
  4f41cde

  Copy full SHA for 4f41cde

• added reward corruption environment

  

  vkrakovna

  authored and

  aslanides

  committed
  c4bf3a6

  Copy full SHA for c4bf3a6

Commits on May 30, 2017

• update experiments

  aslanides

  committed
  7a1daee

  Copy full SHA for 7a1daee

• update plot tool

  aslanides

  committed
  dbf6bc9

  Copy full SHA for dbf6bc9

Commits on May 29, 2017

• update paper title

  aslanides

  committed
  ae70b38

  Copy full SHA for ae70b38

Commits on May 18, 2017

• add (PO)MDP toggle for Q-Learning

  aslanides

  committed
  dd82862

  Copy full SHA for dd82862

Commits on May 11, 2017

• add frac field to experiment API

  aslanides

  committed
  1eec232

  Copy full SHA for 1eec232

Commits on May 9, 2017

• add labels for Q-learning in ipynb

  aslanides

  committed
  9093fd4

  Copy full SHA for 9093fd4

• quality-of-life changes for experiments

  aslanides

  committed
  8b48c83

  Copy full SHA for 8b48c83

Pagination

Previous

Next