AMRL: Aggregated Memory For Reinforcement Learning
https://openreview.net/forum?id=Bkl7bREtDr

Deep Learning vs. Traditional Computer Vision
https://arxiv.org/abs/1910.13796

A radically new cosmological view from a groundbreaking neuroscientist placing the human brain at the center of humanity’s universe. <3

Renowned neuroscientist Miguel Nicolelis introduces readers to a revolutionary new theory of how the human brain evolved to become an organic computer without rival in the known universe. Nicolelis undertakes the first attempt to explain the entirety of human history, culture, and civilization based on a series of recently uncovered key principles of brain function. This new cosmology is centered around three fundamental properties of the human brain: its insurmountable malleability to adapt and learn; its exquisite ability to allow multiple individuals to synchronize their minds around a task, goal, or belief; and its incomparable capacity for abstraction.

Combining insights from such diverse fields as neuroscience, mathematics, evolution, computer science, physics, history, art, and philosophy, Nicolelis presents a neurobiologically based manifesto for the uniqueness of the human mind and a cautionary tale of the threats that technology poses to present and future generations.

Miguel Nicolelis is the Duke School of Medicine Distinguished Professor of Neuroscience and Duke University Professor of Neurobiology, Biomedical Engineering, and Psychology and Neuroscience. In 2004, Scientific American elected him as one of the twenty most influential scientists in the world.

https://www.grexcollc.com/post/recommendation-and-reflections-on-the-true-creator-of-everything-by-miguel-nicolelis

PyTorch internals
http://blog.ezyang.com/2019/05/pytorch-internals/
#ArtificialIntelligence #DeepLearning #Pytorch

TensorBoard in Jupyter Notebooks
Colab Notebook: https://colab.research.google.com/github/tensorflow/tensorboard/blob/master/docs/tensorboard_in_notebooks.ipynb
#TensorFlow #TensorBoard #Jupyter #Notebook

TensorFlow Tutorials
https://www.tensorflow.org/tutorials
#TensorFlow #Tutorials #DeepLearning

The Big Bad NLP Database: Access Nearly 300 Datasets https://www.kdnuggets.com/2020/02/big-bad-nlp-database.html

Neuromodulated Meta-Learning Algorithm)
ANML meta-learns to reduce catastrophic forgetting, and can learn at least 600 tasks sequentially
ANML- Learning to Continually Learn
https://arxiv.org/pdf/2002.09571.pdf

How do we get ready for the Age of AI? What do we need to stay successful?
Being a transformational technology of the digital age, AI is an highly elusive concept and future strategy, involving such critical issues as:

Complexity of its scope and nature; matching AI techniques to problem types; large training data sets; value propositions, the cost and complexity of deploying AI.
https://www.quora.com/How-do-we-get-ready-for-the-Age-of-AI-What-do-we-need-to-stay-successful/answer/Kiryl-Persianov

China:
- We use ConvNets to diagnose COVID19.
- We build a hospital in 10 days.

USA:
- Our hiring freeze left 700 positions at CDC unfilled.
- Our COVID19 test doesn't work.
- We stop our experts from talking to the press.
- The whole pandemic thing is a hoax anyway.

https://www.wired.com/story/chinese-hospitals-deploy-ai-help-diagnose-covid-19/

The camera-ready version and code of my #ICLR2020 paper on building G-CNNS that are equivariant to Lie group transformations

paper B-Spline CNNs on Lie groups
https://openreview.net/forum?id=H1gBhkBFDH
code https://github.com/ebekkers/gsplinets

Introduction to Reinforcement Learning
By DeepMind : https://youtube.com/watch?v=2pWv7GOvuf0&list=PLqYmG7hTraZDM-OYHWgPebj2MfCFzFObQ
#ArtificialIntelligence #DeepLearning #ReinforcementLearning

Brain function relies on circuits of spiking neurons with synapses playing the key role of merging transmission with memory storage and processing. Electronics has made important advances to emulate neurons and synapses and brain-computer interfacing concepts that interlink brain and brain-inspired devices are beginning to materialise. We report on memristive links between brain and silicon spiking neurons that emulate transmission and plasticity properties of real synapses. A memristor paired with a metal-thin film titanium oxide microelectrode connects a silicon neuron to a neuron of the rat hippocampus. Memristive plasticity accounts for modulation of connection strength, while transmission is mediated by weighted stimuli through the thin film oxide leading to responses that resemble excitatory postsynaptic potentials. The reverse brain-to-silicon link is established through a microelectrode-memristor pair. On these bases, we demonstrate a three-neuron brain-silicon network where memristive synapses undergo long-term potentiation or depression driven by neuronal firing rates.
https://www.nature.com/articles/s41598-020-58831-9

Quick tutorial on how to test the online artificial intelligence based coronavirus2019 CT scan detection tool (released Feb 26, 2020):

1. Hit English to provide translation of website:

http://121.40.75.149/znyx-ncov/index#/app/online_try

2. Download ct scan images from covid19 research papers, like this one (take note whether the downloaded image represents covid19 positive cases, as labeled on website):

http://images.rsna.org/index.html…

3. Crop the image in paint, i.e. remove the black sides to the left and right of the image, leaving only picture of lungs in center. (Maybe I'll write up a simple Ui to do this cropping automatically)

4. Load ct scan images in website. (Up to 9 images)

5. Wait a few seconds (On 50mpbs internet connection, may be more waiting time on slower connections)

Results: Look for red boxes in resulting image on right hand side.

6. Optionally, you can also load the same ct images in another ai based pneumonia detection tool I prepared (Feb 9, 2020): https://github.com/JordanMicahBennett/SMART-CT-SCAN_BASED-COVID19_VIRUS_DETECTOR

It is advised that governments subsidize Ct scan price if possible, and advise scan personnel in contact with potential covid19 patients, be properly protected. Notably, China has performed thousands of scans. The control of this virus could reasonably save economy far more than cost of ct scan subsidy in the long run (i.e. help prevent mass shutdown of businesses/schools etc).

Project:

https://github.com/JordanMicahBennett/SMART-CT-SCAN_BASED-COVID19_VIRUS_DETECTOR