Studies in Machine Learning & AI

cyberpunk

Papers

Costs of extinction risk mitigation. A Cost-Benefit Analysis of the extinction risk mitigation, claiming that the annual cost of reducing the probability of human extinction by 0.01% is within the range of $1.1 billion to $3.5 trillion.
Written Memories: Understanding, Deriving and Extending the LSTM.
(Self-titled) The best explanation of Convolutional Neural Networks on the Internet.
Secure Group Testing.
Exploring Facebook’s massive, picture-painting AI brain. A 350,000-square-foot building with Facebook’s most valuable artificial intelligence tools, Big Sur.
Facebook to open-source AI hardware design. Facebook's Open Compute Project GPU hardware contribution, FAIR, which has achieved lots of advancements in the development of AI training hardware.
Software faults raise questions about the validity of brain studies.
Neural correlates of interspecies perspective taking in the post-mortem Atlantic Salmon: An argument for multiple comparisons correction. Researchers show claims showing that the interpretation of functional MRI data is questionable.
Adversarial Machine Learning for Security. A marvelous introduction to AML, with several references.
DeepDrive: self-driving car AI. DNN resources for a self-driving car, including integration with OpenAI Gym.
2016 summer program recap. A Colloquium Series on Robust and Beneficial AI (CSRBAI) at the MIRI office, co-hosted with the Oxford Future of Humanity Institute.
Teaching an AI to write Python code with Python code.

Black Hat 2016: Applied Machine Learning for Data Exfil and Other Fun Topics. Cylance's researchers show some examples of how they apply K-means, Decision Trees, and Markov Chains to security's problems.
Black Hat 2016: An AI approach to Malware Similarity Analysis. Invincea Lab's Researcher shows how they use supervised learning approaches to extract features in Malware data.
Peter Norvig: How Computers Learn. Great introduction and review of AI and ML's history.

cyberpunk