20230505-MTOA: Multitasking agents improve their average and maximum accuracy when tasks overlap.

This archive contains the results of a multi-agent simulation experiment [1] carried out with Lazy lavender [2] environment.<br><br>Experiment Label: 20230505-MTOA<br><br>Experiment design: Agents specialize by accepting or not to play a task.<br><br>Experiment setting: Agents are trained with respect to different tasks and then coordinate upon acting on them. Each time they disagree, one agent adapts its knowledge with respect to the current task.<br><br>Hypotheses: Agents will improve their accuracy more on tasks they choose to play.<br><br>Detailed information can be found in index.html or notebook.ipynb.<br><br>[1] <a href="https://sake.re/20230505-MTOA">https://sake.re/20230505-MTOA</a><br>[2] <a href="https://gitlab.inria.fr/moex/lazylav/">https://gitlab.inria.fr/moex/lazylav/</a><br><br>