Abstract: In this letter, we study the problem of legitimate in-neighborhood learning in multi-agent systems, where stochastic observations of trust between agents are available. Unlike previous works ...