イベント

開催内容

第5回 神戸大学CMDS先端セミナー

タイトル

自然言語処理は深層学習で解けるか：医療言語情報処理からみた今後の課題

講演者

狩野 芳伸 博士（静岡大学 情報学部准教授）

日時

2019年5月6日（月）13:20～14:50

場所

神戸大学工学部 本館2F C2-201教室

概要

自然言語処理分野では、ここ数年深層学習の利用がトレンドであり、最近ではBERTをはじめとするニューラル言語モデルを利用した成果が多く発表されている。一方で、単なるend-to-endの学習では困難な研究課題も多々ある。セミナーでは、最近の深層学習モデルを簡単に紹介し、発表者の研究テーマから主に精神疾患の自動診断や電子カルテの処理など医療言語情報処理を題材として、今後できそうな点、困難な点を議論する。

第4回 神戸大学CMDS先端セミナー

タイトル

Social Networks: "Towards a Visual Inference of Personality Traits based on Images Shared in Social Networks"

講演者

Dr. Jordi Gonzalez（バルセロナ自治大学(UAB)）

日時

2019年4月24日（水）17:00-18:00

場所

神戸大学工学部 本館3F　C2-301教室

概要

The social media, as a major platform for communication and information exchange, provides a rich repository of people's opinions and sentiments about a vast spectrum of topics. Such knowledge is embedded in multiple facets, such as comments, tags, as well as shared image and video content. The analysis of such information either in the area of opinion mining, affective computing or sentiment analysis is playing an important role in computational social sciences, which aims to understand and predict human decision making and enables applications such as brand monitoring, market prediction, and even voting forecasts. However, the massive growth of photo and video sharing is increasingly eclipsing text on the leading visual social platforms. So visual communication is complementing and even supplanting the written word, and this visual language is a powerful way for people to express themselves. This talk will exploit the most recent image understanding models based on neural networks to process the vast amount of data generated by social users. These improvements will enable to know more accurately the social user's demands and cultural-driven interests, eventually reaching some degree of personality trait description, depression detection and early suicidal tendencies estimation.

第3回 神戸大学CMDS先端セミナー

タイトル

Deep Learning: "Going beyond Deep Learning in Understanding Human Behaviors in Image Sequences"

講演者

Dr. Jordi Gonzalez （バルセロナ自治大学(UAB)）

日時

2019年4月23日（火）17:00-18:00

場所

神戸大学工学部 本館3F　C2-301教室

概要

Although the recently re-discovered field of deep learning has revolutionized areas like computer vision, such approaches have still several limitations that should be kept in mind in order to design more advanced artificial general intelligent systems for human behavior understanding and human-computer interaction, among other complex tasks. In particular for those two domains, there indeed is a potentially infinite range of input motions with a potentially infinite range interpretations. In this talk, we will cover how the combination of rule-based symbolic reasoning, prior knowledge and common-sense integration, and high-level abstract concepts manipulation can provide more insights to properly interpret the behaviors of humans in front of a camera. If concepts like space, time and object can be represented, the use of hierarchical symbolic systems used for inference can strongly benefit from the so powerful classification performance achieved by neural networks. As a result, by considering deep learning as just part of a more complex and more challenging problem-solving system, a better understanding of human motion can be achieved, including abstraction, structure, intention, open-world and discovery capabilities.

※ 第１回、第２回（2018年度開催分）の内容はこちらをご覧ください

問い合わせ先
こちらをご参照ください。