No abstract

We consider artificial agents that learn to jointly control their gripper and camera in order to reinforcement learn manipulation policies in the presence of occlusions from distractor objects. Distractors often occlude the object of interest and cause it to disappear from the field of view. We propose hand/eye controllers that learn to move the camera to keep the object within the field of view and visible, in coordination to manipulating it to achieve the desired goal, e.g., pushing it to a target location. We incorporate structural biases of object-centric attention within our actor-critic architectures, which our experiments suggest to be a key for good performance. Our results further highlight the importance of curriculum with regards to environment difficulty. The resulting active vision / manipulation policies outperform static camera setups for a variety of cluttered environments.

No abstract

BackgroundPhylogenetic tree reconciliation is a widely-used method for inferring the evolutionary histories of genes and species. In the duplication-loss-coalescence (DLC) model, we seek a reconciliation that explains the incongruence between a gene and species tree using gene duplication, loss, and deep coalescence events. In the maximum parsimony framework, costs are associated with these event types and a reconciliation is sought that minimizes the total cost of the events required to map the gene tree onto the species tree.ResultsWe show that this problem is NP-hard even for the special case of minimizing the number of duplications. We then show that the problem is APX-hard when both duplications and losses are considered, implying that no polynomial-time approximation scheme can exist for the problem unless P = NP.ConclusionsThese intractability results are likely to guide future research on algorithmic aspects of the DLC-reconciliation problem.