Name: Roland Pihlakas
Type: User
Company: Simplify / Macrotec LLC
Bio: An independent AI safety researcher. I studied psychology, have 19 years of experience in modelling natural intelligence and in designing various AI algorithms.
Location: Estonia
Blog: http://www.simplify.ee/
Roland Pihlakas's Projects
Freedom Of Information Portal
Python package to easily retrain OpenAI's GPT-2 text-generating model on new texts
An implementation of training for GPT2, supports TPUs
Automatically exported from code.google.com/p/hyperdbg
.Net bindings to the V8 JavaScript engine
Automatically exported from code.google.com/p/jonathan-sifs-filesystem
Automatically exported from code.google.com/p/judyarray
The main libseccomp repository
Automatically exported from code.google.com/p/livedump
Automatically exported from code.google.com/p/lzham
A Windows API hooking library
Automatically exported from code.google.com/p/minify
Automatically exported from code.google.com/p/native-nt-toolkit
Functions for overwriting IL of .NET methods
This node module let you see console of other commands.
Curiosity-driven Exploration for Deep Reinforcement Learning by Self-supervised Prediction
Automatically exported from code.google.com/p/omaha
Python implementation of Kernel Canonical Correlation Analysis
Regularized kernel canonical correlation analysis in Python
Pyro 3.x (old version - unmaintained - use Pyro4 instead)
Web-based remote UI for Python's PDB debugger
This tool provide "remote" signtool
Reproducing the results of "Playing Atari with Deep Reinforcement Learning" by DeepMind
Reproducing 'Gridworlds agents, pycolab examples
converts data from RiigiTeataja xml-s to own database
Test DRAM for bit flips caused by the rowhammer problem
Rowhammer.js - A Remote Software-Induced Fault Attack in JavaScript