Page Not Found
Page not found. Your pixels are in another canvas.
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Page not found. Your pixels are in another canvas.
My Github hosted projects
A Model-Predictive-Control style list of ML tasks to do
Fastai transforms don’t seem to help DonkeyCar training
Dropout layers make VAE output worse
My speculation on how Avengers Endgame could go
Table of algorithms in the Stable Baselines repo
Using DAgger to improve MaLPi’s training, while MaLPi is driving
Using DAgger to improve MaLPi’s training
One of MaLPi’s first fully autonomous laps
Hyperparameter optimization using hyperopt on racetrack data.
Testing image normalization when using a GRU.
Testing image normalization when using an LSTM.
I had completely forgotten to normalize the images I’m feeding into MaLPi’s network, so I thought I’d try to be a bit more formal about it than my usual.
Training on batch sizes and/or sequence lengths longer than one, while still being able to run one image at a time on the robot.
I forked Open AI’s baseline code and made a few changes. This was my first full run before I started playing around with the model architecture.
Eliminated a battery! (maybe)
I’m about to make large changes to MaLPi so I wanted to document the current state of the hardware.
What have I gotten myself into?!?! Motors and controllers and a breadboard, oh my!
Some progress on the hardware front.
Test how long the PowerGen battery can run MaLPi on a single charge.
MaLPi (Machine Learning Pi)
Using RL value functions to encode semantic knowledge, specifically by a robot.
Off-Policy AC with linear state features. Includes elegibility traces.
One of the first deep reinforcement learning papers.
Improved Q-value estimation by reducing overestimates of Deep Q-networks.
In-progress second edition of an RL textbook.
Overview (with references) of attention and several types of augmentation for RNNs.
Increase speed of a Reinforcement Learning system with auxiliary task.
Five questions to ask about your deep learning project.
Relationships between objects.
A single ML model used for very different tasks.
Notes on using Eligibility Traces with neural networks
A long review of the use of DL in robotics
A long review of the use of DL in robotics
Pre-train using supervised learning on human provided demonstations.
AlphaGo Zero, all RL self-play.
Tensorizing LSTMs to make them wider and deeper without adding parameters and with minimal extra compute costs.
Pretraining Deep Actor-Critic Reinforcement Learning Algorithms With Expert Demonstrations.
Unsupervised learning of image encoding and dynamics model.
Unsupervised training of a memory that is used for prediction of state and reward.
Unsupervised learning of image encoding, dynamics and reward models.
Learned dynamics model with a GAN for image generation and MCTS for planning.
Continual learning with a universal, off-policy agent.
Access consciousness and it’s relation to general intelligence