7 accepted papers
In gradient-based learning, a step size chosen in parameter units does not produce a predictable per-step change in the function output. This may lead to instability in the streaming setting (i.e., batch size=1), where stochasticity is not averaged out and update magnitudes can momentarily become ar…
Deep representation learning methods struggle with continual learning, suffering from both catastrophic forgetting of useful units and loss of plasticity, often due to rigid and unuseful units. While many methods address these two issues separately, only a few currently deal with both simultaneously…
Modern deep policy gradient methods achieve effective performance on simulated robotic tasks, but they all require large replay buffers or expensive batch updates, or both, making them incompatible for real systems with resource-limited computers. We show that these methods fail catastrophically whe…
Second-order information is valuable for many applications but challenging to compute. Several works focus on computing or approximating Hessian diagonals, but even this simplification introduces significant additional costs compared to computing a gradient. In the absence of efficient exact computa…
The optoelectronic microrobot is an advanced light-controlled micromanipulation technology which has particular promise for collecting and transporting sensitive microscopic objects such as biological cells. However, wider application of the technology is currently limited by a reliance on manual co
Optoelectronic tweezer-driven microrobots (OETdMs) are a versatile micromanipulation technology based on the application of light induced dielectrophoresis to move small dielectric structures (microrobots) across a photoconductive substrate. The microrobots in turn can be used to exert forces on sec…
Interaction is fundamental in autonomous driving (AD). Despite more than a decade of intensive R&D in AD, how to dynamically interact with diverse road users in various contexts still remains unsolved. Multi-agent learning has recently seen big breakthroughs and has much to offer towards solving rea