configuration_multiplayers_with_aggregation module

Configuration for the simulations, for multi-players with aggregation.

configuration_multiplayers_with_aggregation.HORIZON = 10000

HORIZON : number of time steps of the experiments. Warning Should be >= 10000 to be interesting “asymptotically”.

configuration_multiplayers_with_aggregation.REPETITIONS = 200

REPETITIONS : number of repetitions of the experiments. Warning: Should be >= 10 to be statistically trustworthy.

configuration_multiplayers_with_aggregation.DO_PARALLEL = True

To profile the code, turn down parallel computing

configuration_multiplayers_with_aggregation.N_JOBS = -1

Number of jobs to use for the parallel computations. -1 means all the CPU cores, 1 means no parallelization.

configuration_multiplayers_with_aggregation.NB_PLAYERS = 6

NB_PLAYERS : number of players for the game. Should be >= 2 and <= number of arms.

configuration_multiplayers_with_aggregation.collisionModel(t, arms, players, choices, rewards, pulls, collisions)

The best collision model: none of the colliding users get any reward

configuration_multiplayers_with_aggregation.VARIANCE = 0.05

Variance of Gaussian arms

configuration_multiplayers_with_aggregation.CACHE_REWARDS = False

Should we cache rewards? The random rewards will be the same for all the REPETITIONS simulations for each algorithms.

configuration_multiplayers_with_aggregation.NB_ARMS = 12

Number of arms for non-hard-coded problems (Bayesian problems)

configuration_multiplayers_with_aggregation.LOWER = 0.0

Default value for the lower value of means

configuration_multiplayers_with_aggregation.AMPLITUDE = 1.0

Default value for the amplitude value of means

configuration_multiplayers_with_aggregation.ARM_TYPE

alias of Arms.Bernoulli.Bernoulli

configuration_multiplayers_with_aggregation.ENVIRONMENT_BAYESIAN = False

True to use bayesian problem

configuration_multiplayers_with_aggregation.MEANS = [0.05, 0.1318181818181818, 0.21363636363636362, 0.2954545454545454, 0.3772727272727272, 0.459090909090909, 0.5409090909090909, 0.6227272727272727, 0.7045454545454545, 0.7863636363636363, 0.868181818181818, 0.95]

Means of arms for non-hard-coded problems (non Bayesian)

configuration_multiplayers_with_aggregation.configuration = {'averageOn': 0.001, 'change_labels': {0: 'Aggr(rhoRand, RandTopM, MCTopM, Selfish) kl-UCB', 1: 'rhoRand kl-UCB', 2: 'RandTopM kl-UCB', 3: 'MCTopM kl-UCB', 4: 'Selfish kl-UCB', 5: 'Centralized kl-UCB'}, 'collisionModel': <function onlyUniqUserGetsReward>, 'environment': [{'arm_type': <class 'Arms.Bernoulli.Bernoulli'>, 'params': [0.05, 0.1318181818181818, 0.21363636363636362, 0.2954545454545454, 0.3772727272727272, 0.459090909090909, 0.5409090909090909, 0.6227272727272727, 0.7045454545454545, 0.7863636363636363, 0.868181818181818, 0.95]}], 'finalRanksOnAverage': True, 'horizon': 10000, 'n_jobs': -1, 'players': [Selfish(UCB), Selfish(UCB), Selfish(UCB), Selfish(UCB), Selfish(UCB), Selfish(UCB)], 'plot_lowerbounds': False, 'repetitions': 200, 'successive_players': [[<Policies.Aggregator.Aggregator object>, <Policies.Aggregator.Aggregator object>, <Policies.Aggregator.Aggregator object>, <Policies.Aggregator.Aggregator object>, <Policies.Aggregator.Aggregator object>, <Policies.Aggregator.Aggregator object>], [rhoRand(Aggregator($N=3$)), rhoRand(Aggregator($N=3$)), rhoRand(Aggregator($N=3$)), rhoRand(Aggregator($N=3$)), rhoRand(Aggregator($N=3$)), rhoRand(Aggregator($N=3$))], [RandTopM(Aggregator($N=3$)), RandTopM(Aggregator($N=3$)), RandTopM(Aggregator($N=3$)), RandTopM(Aggregator($N=3$)), RandTopM(Aggregator($N=3$)), RandTopM(Aggregator($N=3$))], [MCTopM(Aggregator($N=3$)), MCTopM(Aggregator($N=3$)), MCTopM(Aggregator($N=3$)), MCTopM(Aggregator($N=3$)), MCTopM(Aggregator($N=3$)), MCTopM(Aggregator($N=3$))], [Selfish(Aggregator($N=3$)), Selfish(Aggregator($N=3$)), Selfish(Aggregator($N=3$)), Selfish(Aggregator($N=3$)), Selfish(Aggregator($N=3$)), Selfish(Aggregator($N=3$))], [CentralizedMultiplePlay(Aggregator($N=3$)), CentralizedMultiplePlay(Aggregator($N=3$)), CentralizedMultiplePlay(Aggregator($N=3$)), CentralizedMultiplePlay(Aggregator($N=3$)), CentralizedMultiplePlay(Aggregator($N=3$)), CentralizedMultiplePlay(Aggregator($N=3$))]], 'verbosity': 6}

This dictionary configures the experiments

configuration_multiplayers_with_aggregation.nbArms = 12

Number of arms in the first environment