Collaborating Authors

Topic Modeling with Wasserstein Autoencoders Artificial Intelligence

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.

Iran claims it downed 'unknown' drone over Persian Gulf, Pentagon says all US devices accounted for

FOX News

Ayatollah Khamenei doubles down on Iran's commitment not to engage in talks with the United States; Trey Yingst reports. The drone was reportedly hit in the early morning at the port city of Mahshahr, which is in the oil-rich Khuzestan province and lies on the Persian Gulf. "The downed droned definitely belonged to a foreign country. Its wreckage has been recovered and is being investigated," the governor of Khuzestan, Gholamreza Shariati, said, according to the official IRNA news agency. He said the drone violated Iran's airspace but did not provide any additional information, including whether it was a military or civilian drone.

Erica Mena Talks Leaving Bow Wow, Says His Current Level Of Fame Bothers Him

International Business Times

Since their split in 2015, Erica Mena and Bow Wow have sent several shots at each other on social media, and now the former "Love & Hip-Hop" star is telling people exactly how she feels about him. In an interview with VladTV, Mena said although she tries to ignore her ex's posts, she still feels sorry for him at times. "Since leaving that situation, it doesn't really get attention from me or a reaction" said the 28-year-old. "I'll hear or see under my hashtag how he's re-posting a list of women he's been with and bragging about it ... and I just feel sorry … It's annoying to wake up one day to see pictures of me that he posts on his Facebook to try and make you guys believe that we're still together. I'm always thrown off by him," added Mena.

Bow Wow Blasts Erica Mena On Social Media For Lying; Says He Actually Broke Up With Her

International Business Times

Did Erica Mena leave Bow Wow or was it the other way around? That seems to be the current debate going on between the former couple after the New York model spoke about the breakup in a recent interview. According to Bow Wow, 29, his ex-fiancée was lying about ending the relationship, and he said there were witnesses around when they split. "Even the homies was there that day when it went down. That's between us, though," he wrote on VladTV's Instagram page, since VladTV is where Mena, 28, gave her interview.

Semantic Hypergraphs Artificial Intelligence

Existing computational methods for the analysis of corpora of text in natural language are still far from approaching a human level of understanding. We attempt to advance the state of the art by introducing a model and algorithmic framework to transform text into recursively structured data. We apply this to the analysis of news titles extracted from a social news aggregation website. We show that a recursive ordered hypergraph is a sufficiently generic structure to represent significant number of fundamental natural language constructs, with advantages over conventional approaches such as semantic graphs. We present a pipeline of transformations from the output of conventional NLP algorithms to such hypergraphs, which we denote as semantic hypergraphs. The features of these transformations include the creation of new concepts from existing ones, the organisation of statements into regular structures of predicates followed by an arbitrary number of entities and the ability to represent statements about other statements. We demonstrate knowledge inference from the hypergraph, identifying claims and expressions of conflicts, along with their participating actors and topics. We show how this enables the actor-centric summarization of conflicts, comparison of topics of claims between actors and networks of conflicts between actors in the context of a given topic. On the whole, we propose a hypergraphic knowledge representation model that can be used to provide effective overviews of a large corpus of text in natural language.