Topic: visual inputs

Nikita Rudin discusses reinforcement learning for robot locomotion and the difficulties of transferring policies from simulation to physical robots. Topics include visual inputs, real-world data for improving simulators, imitation learning and teleoperation, hierarchical control, and trade-offs between modular and end-to-end approaches for autonomous robots.

Visual working memory (VWM) allows us to maintain a small amount of task-relevant information in mind so that we can use them to guide our behavior. Although past studies have successfully characterized its capacity limit and representational quality during maintenance, the consequence of its usage for task-relevant behaviors has been largely unknown. In this talk, I will demonstrate that VWM representations get distorted when they are used for perceptual comparisons with new visual inputs, especially when the inputs are subjectively similar to the VWM representations. Furthermore, I will show

Background: Many animals use celestial cues to determine their heading. In the insect brain, the central complex is crucial for spatial orientation. In desert locusts and fruit flies, compass neurons in the central complex encode the animal’s heading relative to the sun [2, 5]. However, the neural computations underlying the transformations of visual inputs to a compass signal remain elusive. In fruit flies, it is hypothesised that visual information is relayed from tangential neu- rons to compass neurons through likely inhibitory all-to-all connections with plastic synapses [1]. We aim to exp