We consider the problem of distributed throughput maximization for multi-channel ALOHA networks. We focus on networks containing a large number of users that transmit over a low number of channels. First, we consider the problem of constrained distributed rate maximization, where user rates are subject to total transmission probability constraints. We propose a distributed best-response algorithm to solve the rate maximization problem, where each user updates its strategy using its local channel state information (CSI) and by monitoring the channel utilization. We then consider the case where users are not restricted by transmission probability constraints. Distributed optimization of the network throughput under uncertainty is mandatory since the transmission probabilities of other users are unknown. We propose a distributed scheme to solve the throughput optimization problem under uncertainty, where users adjust their transmission probability to maximize their rates, but maintain the desired load on the channels. We propose sequential and parallel algorithms for this purpose.