Twitter | Search | |
This is the legacy version of twitter.com. We will be shutting it down on 15 December 2020. Please switch to a supported browser or device. You can see a list of supported browsers in our Help Center.
Search Refresh
Jamal J8 🕸 Nov 15
R2 Day 36: BIG BIG DAY. Finished the basic Gradient Descent algo. Tuned the learning_rate and finally get why it needs tuning based on the aglo. PROGRESS 💪 Big Step 🐢
Reply Retweet Like
Jamal J8 🕸 Nov 21
R2 Day 42: Reviewed my work so far. Started working on the gradient descent algo for two variables -> 3D 😱. So close. Yet so far. Also designed 👇based on a Tutorial
Reply Retweet Like
Jamal J8 🕸 10h
R2 Day 44: Writing the gradient descent algorithm for both variables. Still need to figure out how and why the weighting works. Step by step 🐢
Reply Retweet Like
Jason Normanton 🤍💙💛 48m
Adversarial Attacks on Deep Neural Networks
Reply Retweet Like
Antonio Grasso Nov 19
What are the main differences between a biological neuron and a neuron simulated by computer neural networks? Huge, uncomparable. Yet, artificial intelligence will smartly help us to be even more human. by
Reply Retweet Like
Vishal Panchal Nov 17
on Sequence Data — Tutorial on creating Gated Recurrent Units (GRU) in Recurrent Neural Networks (RNN) with : —— v/
Reply Retweet Like
Jamal J8 🕸 Nov 16
R2 Day 37: Ran into some bugs.. After 89 loops the descent stopped at a value much higher than the goal. I have some ideas for the solution. Also the COLORWAYS😁
Reply Retweet Like
Jamal J8 🕸 Nov 22
R2 Day 43: It feels great when everything starts to come together. Partial derivatives and building the algo for the two variables. Step by Step 🐢 Is my answer right?? 👇
Reply Retweet Like
Jamal J8 🕸 Nov 18
R2 Day 39: Long day... Got the chain rule down. Converted formula to the derivative. Also tried to update Icy for a new game. Maybe. Not confirmed yet. Thoughts?
Reply Retweet Like
Marcus Borba Nov 15
Reply Retweet Like
Jamal J8 🕸 Nov 19
R2 Day 40: Ran the descent focused on the derivatives but the same problem occurs. The descent stops short of the goal. I'm kinda frustrated. 😠Work you damn code you!🤣
Reply Retweet Like
Marcus Borba Nov 17
Reply Retweet Like
Dr. Marcell Vollmer #StayHome #StaySafe Nov 16
This is what Autopilot sees using that take 70.000 GPU hours to train and output 1,000 tensors (predictions) at each timestep. MT
Reply Retweet Like
PINAKI Nov 22
Reply Retweet Like
Aditics 36m
What is Keras? The deep neural network API explained via
Reply Retweet Like
Computational Systems Neuroscience Nov 18
Press release: Nervous systems of insects inspire efficient future systems on our recent paper on sensory processing, biologically inspired and sensorimotor control with spiking :
Reply Retweet Like
Bering Nov 22
It's really easy to introduce I/O mistakes to experiments. We added a simple method to the datasets package that allows you to take a peak at data input/processing/augmentation results before training your . Check it out:
Reply Retweet Like
Kirk Borne Nov 16
Reply Retweet Like
Domesticated Brain Nov 22
Reply Retweet Like
Dr. Sally Eaves Nov 21
Reply Retweet Like