Most existing lifelong machine learning works focus on how to exploit previously experiences (i.e., knowledge library) from current tasks, and transfer it to learn a new task. However, when encountering with a large pool of candidate tasks, the knowledge among various coming tasks are imbalance, and the system should intelligently choose the next task to learn. In this paper, an effective human cognition strategy is taken into consideration, and preferentially selecting the most valuable task to learn. More specifically, we assess the importance of the new coming task as an outlier detection issue, and propose to employ a watchdog knowledge library to reconstruct each task under l0 -norm sparse constraint. The coming candidate tasks are then sorted depending on the sparse reconstruction score in a descending order, which is referred to as watchdog mechanism. Following this, we design a hierarchical knowledge library for the lifelong learning framework to encode new task with higher reconstruction score, where the library consists of two-level task descriptors: a high-dimensional one with low-rank constraint and a low-dimensional one. Experimental results on several existing benchmarks demonstrate that our proposed model outperforms several state-of-the-arts.