It was the final week of my project, and I’m actually genuinely pretty sad to see it go. I started out by trying to improve the objective function, as promised. I did manage to make it so that it was slightly more accurate on the testing data; unfortunately for me, this did not really change the quality of the generated sentences. I tried tweaking the design a lot, but in the end, the objective function was a limiting factor. I’ve come to the conclusion over this project that maybe I should have implemented it with a GAN; the problem is that so many design iterations passed that by the time that was clear, it was a little bit too late for that.
My final coding-related task was to compare my system with random noise. To that end, I generated the same number of sentences using completely random noise. It was reassuring to see that those sentences were categorically worse than the sentences generated by my system. From a quality diversity standpoint, the solutions generated by the network scored much higher than the scores for the random sentences, which indicates that the system was successful.
After that, my main focus was on writing my final report. A lot of that was trying to wrangle overleaf, and much of it was also going through the literature to try to find the exact sources I needed. Ultimately, I think I’m pretty happy with the final results.
As I mentioned, I am sad that this is over. I gained a lot of technical skills, and between this and preparing for my thesis I feel far more prepared for grad school than I did before. Obviously, I owe a lot to Nathan and to Dr. Mataric for mentoring me, and I’m really glad I decided to do DREU.