Augmented Random Search is a game changing AI. It is very simple implementation it is able to do the exactly the same thing that Google Deepmind did in there accomplishment last year which is to train an AI to work and run across a field. Augmented Random Search is 100x time faster and 100x time more powerful.
The best part is that with augmented random search is no need for complex Algorithms or Framework such as Tensorflow or PyTorch. This is the best AI in 2018.
A common belief in model-free reinforcement learning is that methods based on random search in the parameter space of policies exhibit significantly worse sample complexity than those that explore the space of actions. We dispel such beliefs by introducing a random search method for training static, linear policies for continuous control problems, matching state-ofthe-art sample efficiency on the benchmark MuJoCo locomotion tasks. Our method also finds a nearly optimal controller for a challenging instance of the Linear Quadratic Regulator, a classical problem in control theory, when the dynamics are not known. Computationally, our random search algorithm is at least 15 times more efficient than the fastest competing model-free methods on these benchmarks. We take advantage of this computational efficiency to evaluate the performance of our method over hundreds of random seeds and many different hyperparameter configurations for each benchmark task. Our simulations highlight a high variability in performance in these benchmark tasks, suggesting that commonly used estimations of sample efficiency do not adequately evaluate the performance of RL algorithms.
https://arxiv.org/pdf/1803.07055.pdf
$ pip install -r requirements.txt
$ python ars.py