A set of scripts used in my PhD dissertation. It allows to design Schroeder acoustic diffusers with use of deep reinforcement learning, genetic algorithms, and random design process. The design process employing reinforcement learning and genetic algorithms is driven by a coupled physical simulation of acoustic waves propagation based on the finite-difference time-difference (FDTD) algorithm.
If you find this code useful, I would be very glad if you cite this publication:
A. Kurowski and B. Kostek, "Reinforcement Learning Algorithm and FDTD-Based Simulation Applied to Schroeder Diffuser Design Optimization," in IEEE Access, vol. 9, pp. 136004-136017, 2021, doi: 10.1109/ACCESS.2021.3114628.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.