Stuff I find to be a interesting read... that discusses some novel ideas in mathamatics, neuroscience, machine learning, or the intersection in between. (My notes on literatures)
Lakatosian research program setting computational understanding of the brain.
Building Machines that Learn and Think like PeopleSome foundation review that discusses potential research directions in how we can progress in AI for making more human-like machines.
Probabilistic flavour infused cutting-edge actor-critic.
Proximal Policy Optimization AlgorithmsActor-critic algorithm on steriods.
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner ArchitecturesDistributed training training framework of RL algorithm to perform off-policy learning.
Addiction as a Computational Process Gone AwryReinforcement Learning methods (TDRL) highly resembles physical neuronal mechanism greatly. Now using it to model the adiction process.
One of the earliestwork in proposing using ANN to model the brain.
Deep Neuroethology of a Virtual RodentOne of the earliestwork in aligning deep reinforcement learning with a bilogical counterpart.
Whole-body Simulation of Realistic Fruit Fly Locomotion with Deep Reinforcement LearningDistributed trained MPO policy for goal-directed RL and imitation learning for flybody model.
Done by our collaborator's lab at Harvard, using imitation learning to mimic the behaviours of real rodent and shown similar neuronal activations across natural and artificial agents.
CoMic: Complementary Task Learning & Mimicry for Reusable SkillsUsing encoder/decoder architecture to transfer motor skills across task and build a low level controller.
Building beliefs in artificial agent through a MOPDP condition.