Distributed multi-task learning provides significant advantages in multi-agent networkswithheterogeneous datasources where agents aimtolearndistinctbut correlated models simultaneously. However, distributed algorithms for learning relatedness among tasks arenotresilient inthepresence ofByzantine agents. In this paper, we present an approach for Byzantine resilient distributed multi-task learning. We propose an efficient online weight assignment rule by measuring the accumulated loss using an agent's data and its neighbors' models. A small accumulated loss indicates a large similarity between the two tasks.

Weconsider Diplomacy,a7-player board game designed toaccentuate dilemmas resulting from many-agent interactions.

We prove the existence of this geometry for a wide class of real world games by exposing their temporal nature.

Then we can update the joint distribution for the L.H.S. withϱl = ϱ 12 by exchanging the22 We prove the triangle inequality by contradictions similar to iii). Each agent has to resolve to select the action from its discrete action space to move around. Themixingnetwork56 has one hyper-layer as described in QMIX with64 units. The optimizer to optimize the neural57 networks is "Adam". Each URL algorithm is deployed to learn different joint policies (Z = 1058 for MPE andZ = 20 for GRF) and mixing networks every time.