System, method and product for task allocation

A computerized apparatus having a processor, the processor being adapted to perform the steps of:

• obtaining a stream of tasks in a system, wherein the tasks are to be performed by agents, wherein the agents choose whether or not to perform a task, wherein a quality of each agent affects contributions of performing tasks by the agent;
• calculating for each agent, an average quality of tasks in the system that were completed in the past by the agent;
• allocating the tasks to the agents, wherein said allocating comprises selecting an agent to perform a task, wherein said selecting is based on the average quality of the agent, wherein said allocating comprises applying a probabilistic regret minimization algorithm to select agents to perform the tasks, wherein a probability to select agent i by the probability regret minimization algorithm is based on the following formula:
• p
• i
• =
• (
• 1
• -
• γ
• )
• ·
• ⅇ
• q
• i
• ∑
• j
• =
• 1
• n
• ⁢
• ⅇ
• q
• j
• +
• γ
• ·
• 1
• n
• ,
• wherein γ is a value between zero and one, wherein qi is the average quality associated with agent i, wherein n is a number of agents from which the probability minimization algorithm is configured to select;
• obtaining an indication that the agent has completed the task;
• in response to the indication, computing a reward for the agent, wherein the reward is calculated according to a total contribution of the agent to the system by completing the task;
• outputting the reward to be provided to the user;
• whereby biasing said allocating to prefer allocating tasks to a first agent over a second agent, if a quality of the first agent is greater than a quality of the second agent, wherein said biasing is not dependent on prior knowledge of the qualities.

A computer program product comprising a computer readable storage medium retaining program instructions, which program instructions when read by a processor, cause the processor to perform a method comprising:

• obtaining a stream of tasks in a system, wherein the tasks are to be performed by agents, wherein the agents choose whether or not to perform a task, wherein a quality of each agent affects contributions of performing tasks by the agent;
• calculating for each agent, an average quality of tasks in the system that were completed in the past by the agent;
• allocating the tasks to the agents, wherein said allocating comprises selecting an agent to perform a task, wherein said selecting is based on the average quality of the agent, wherein said allocating comprises applying a probabilistic regret minimization algorithm to select agents to perform the tasks, wherein a probability to select agent i by the probability regret minimization algorithm is based on the following formula:
• p
• i
• =
• (
• 1
• -
• γ
• )
• ·
• ⅇ
• q
• i
• ∑
• j
• =
• 1
• n
• ⁢
• ⅇ
• q
• j
• +
• γ
• ·
• 1
• n
• ,
• wherein γ is a value between zero and one, wherein qi is the average quality associated with agent i, wherein n is a number of agents from which the probability minimization algorithm is configured to select;
• obtaining an indication that the agent has completed the task;
• in response to the indication, computing a reward for the agent, wherein the reward is calculated according to a total contribution of the agent to the system by completing the task;
• outputting the reward to be provided to the user;
• whereby biasing said allocating to prefer allocating tasks to a first agent over a second agent, if a quality of the first agent is greater than a quality of the second agent, wherein said biasing is not dependent on prior knowledge of the qualities.