We consider a new variant of the multi-robot task allocation problem – Inverse Risk-sensitive Multi-Robot Task Allocation (IR-MRTA). “Forward” MRTA – the process of deciding which robot should perform a task given the reward (cost)-related parameters, is widely studied in the multi-robot literature. In this setting, the reward (cost)-related parameters are assumed to be already known: parameters are first fixed offline by domain experts, followed by coordinating robots online. What if these parameters need to be changed during runtime by non-expert human supervisors? This may happen for example, when the human supervisor’s perception of the allocation risk (e.g., the probability of robot failure) changes. In such cases, the robots need to change the parameters of the originally posed task allocation problem based on evolving human preferences. We study such problems through the lens of inverse task allocation, i.e., the process of finding parameters given solutions to the problem. Specifically, we propose a new formulation IR-MRTA, whose goal is to find a new set of parameters of the human behavioral risk model that minimally deviates from the current MRTA parameters while causing a greedy task allocation algorithm to allocate robot resources in line with the updated human supervisory input. We show that even in the simple case this is a non-convex optimization problem. We propose a Branch \(\&\) Bound algorithm (BB-IR-MRTA) to solve such problems. We demonstrate BB-IR-MRTA on numerical simulations of multi-robot target capture and show that it provides significant advantages in running time and peak memory usage compared to a brute-force baseline.