Differentiable Neural Computers for in silico molecular design: Benchmarks of architectures in generative modeling of molecules
Typ
Examensarbete för masterexamen
Program
Publicerad
2019
Författare
PRYKHODKO, OLEKSII
JOHANSSON, SIMON
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
In the area of in silico drug discovery, deep learning has grown immensely as a
field of research. Recurrent neural networks (RNN) is one of the most common
approaches used for generative modeling, but recently the differentiable neural computer
(DNC) has been shown to give considerable improvement over the RNN for
modeling of sequential data. In this thesis, a DNC has been implemented as an
extension to REINVENT, an RNN based model that has already been successfully
shown to generate molecules with high validity. The model was benchmarked on
its capacity to learn the SMILES language on the GDB-13 and ChEMBL datasets.
The DNC shows some improvement on all tests conducted at the cost of greatly
increased computational time and memory consumption, which puts its practical
use into question. This project also gives some insight into the effect of the DNC
hyperparameters for the task of generative modeling of molecules.
Beskrivning
Ämne/nyckelord
Computer science , machine learning , recurrent neural networks , GRU , LSTM , differentiable neural computer , engineering , project , thesis