We propose a novel neural topic model in the Wasserstein autoencoders (WAE) framework. Unlike existing variational autoencoder based models, we directly enforce Dirichlet prior on the latent document-topic vectors. We exploit the structure of the latent space and apply a suitable kernel in minimizing the Maximum Mean Discrepancy (MMD) to perform distribution matching. We discover that MMD performs much better than the Generative Adversarial Network (GAN) in matching high dimensional Dirichlet distribution. We further discover that incorporating randomness in the encoder output during training leads to significantly more coherent topics. To measure the diversity of the produced topics, we propose a simple topic uniqueness metric. Together with the widely used coherence measure NPMI, we offer a more wholistic evaluation of topic quality. Experiments on several real datasets show that our model produces significantly better topics than existing topic models.
While our exposure to technology is often more novel than anything -- like a program that recognizes your friends and tags them in a picture -- small tweaks may turn the mundane into something life-changing. Intel is partnering with Israeli startup MobileODT, a company that created a smartphone app to diagnose cervical cancer. Together, both businesses are searching for (and challenging) developers to arrive at an algorithm that, based on images, accurately identifies women's cervix types. If successful, the effort could save the lives of millions of women around the world who don't have access to adequate medical care. Saul Singer's book Start-up Nation inspired the view of Israel as an enigma in innovation, and around the world, people are looking to learn from Israel in its attempts to construct better tech ecosystems.