Hi! I’m Anthony Francis, and I design intelligent machines and emotional robots, focusing on deep reinforcement learning for indoor robot navigation. My original research page http://www.dresan.com/research/ details my thesis work on contextual memory. More recently, I’ve focused on robotics, first emotional robotics and later robot navigation. More information can be found in the subpages below:
I am the founder of Logical Robotics, a consulting firm specializing in the design and construction of skilled, interpretable, social and efficient intelligent agents. My research involves organizing and promoting the Embodied AI Workshop, the premiere workshop exploring the science of machine learning applied to interactive worlds, and social robot navigation work such as the Social Navigation Symposium.
Until recently, I was a researcher (bio page, archived) for Robotics at Google, focusing on robotics for reinforcement learning and reinforcement learning for robotics. My work there involved making robot navigation more reliable using approaches that combine deep reinforcement learning, evolutionary computation, and “traditional” probabilistic motion planning approaches which were the snazzy new thing when I started my PhD work, and which still do the job pretty well now.
My PhD research focused on contextual memory: how implicit context helps us retrieve and use information more appropriately. This can include not just memory for facts, but memory for emotion: annotating experiences with the emotions they cause can be used to learn emotional responses that are appropriate for different situations.
My other research interests include programming language design, interactive fiction, natural language understanding, animal cognition, cognitive science, intelligent agents, and physics, particularly general relativity. For more information, check out the subpages, or read recent posts below:
Posts About Intelligence … Broadly Construed 🙂
[twenty twenty-four day one three two]: what?!
There’s an ongoing debate over whether human emotions are universal: I, like many researchers, think that there was solid work done by Ekman back in the day that demonstrated this…
[twenty twenty four day one two five]: this is it
This is it. Today, the 4th, is the last day to submit papers to the Embodied AI Workshop 2024, and we are not going to extend this deadline because we’ve gotten…
[twenty twenty-four day one two four]: last call for embodied ai papers!
Hey folks! Today (Saturday May 4th) is the last day to submit papers to the Embodied AI Workshop 2024! Please send us what you’ve got! Just between you and me…
[twenty twenty-four day forty-three]: neurodivergence by the numbers
So! While working on The Neurodiversiverse I’ve been reading up a lot on neurodiversity. According to Devon Price’s Unmasking Autism, autism is massively undiagnosed, and for good—well, understandable—reasons. From parents…
Robots!
Still at the Conference on Robot Learning. LOTS of robot dogs were about, lots of diffusion model and transformer work, and lots of language model planning. More later, gotta crash.…
Announcing Logical Robotics
So, I’m proud to announce my next venture: Logical Robotics, a robot intelligence firm focused on making learning robots work better for people. My research agenda is to combine the…
phewww ….
… finally, a chance to catch a break. It’s been a difficult few weeks due to “the Kerfluffle” which I hope to blog about shortly (those on my LinkedIn have…
Ugh, WordPress updates edition …
… the block editor of WordPress seems to be making my old non-block-editor posts turn into solid walls of text. See the post “Pascal’s Wager and Purchasing Parsley“: Yeah, it’s…
Once again, I’m running deep learning on a Macbook …
… and the trick to getting it working was, as usual, “working just a little bit harder than you want to“. Shortly after my last post, I got REINFORCE, a…
The Embodied AI Workshop is Tomorrow, Sunday, June 20th!
What happens when deep learning hits the real world? Find out at the Embodied AI Workshop this Sunday, June 20th! We’ll have 8 speakers, 3 live Q&A sessions with questions…
