新任助教講演会(Lectures from New Assistant Professors)

日時(Datetime) 令和2年6月16日(火)3限 (13:30 -- 15:00), 2020/06/16, Tuesday, 3rd slot
場所(Location) WebEx
司会(Chair) 張 任遠 (Renyuan Zhang)

講演者(Presenter) Kim Young Woo, 情報セキュリティー工学研究室 (Information Security Engineering Lab.)
題目(Title) A Novel Statistical Eye-diagram Estimation Method in High-speed Digital Systems
概要(Abstract) Recent technology trend requires TB/s bandwidth. To support this trend, data rate/pin in memories’ high-speed channels are continuously increasing. Therefore, to maintain TB/s bandwidth, maintaining signal integrity in the high-speed channel is mandatory. In the high-speed channel, not only channel parameters but also non-linear power/ground noise generated by simultaneous switching buffer outputs (SSOs) noise affect signal integrity. These noises are dominated by the occurrence probabilities of the SSO buffer combinations. However, conventional transient simulators and eye-diagram estimation methods fail to accurately estimate the eye-diagram considering these noise effects. In this presentation, a novel statistical eye-diagram method which considers non-linear power/ground noise and SSO noise is proposed. Four different output responses are derived: pull-up/down and steady states ‘one’ and ‘zero’. These responses are affected by the power/ground noise and SSO noise. Also, formula which derives occurrence probability of each response in function on aggressor buffer states is proposed. By mapping occurrence probabilities to the output response-sets, we can derive PDF of each responses-set. By defining main-cursors/ISI PDFs and taking convolution between main-cursor PDF and ISI PDF which affects each other, we can derive statistical eye-diagram considering impacts of power/ground noise and SSO noise. The proposed method is validated using HSPICE at the BER level where transient simulators can accurately estimate the eye-diagram. The proposed method is fast and accurate. The proposed method is applied for signal integrity analysis in HBM which is memory in form of 3D IC.

講演者(Presenter) 日永田 智絵(Chie Hieida), 数理情報学学研究室 (Mathematical Informatics Lab.)
題目(Title) Deep Emotion: A Computational Model of Emotion Using Deep Neural Networks
概要(Abstract) Emotions are very important for human intelligence. For example, emotions are closely related to the appraisal of the internal bodily state and external stimuli. This helps us to respond quickly to the environment. Another important perspective in human intelligence is the role of emotions in decision-making. Moreover, the social aspect of emotions is also very important. Therefore, if the mechanism of emotions were elucidated, I could advance toward the essential understanding of our natural intelligence. In this study, a model of emotions is proposed to elucidate the mechanism of emotions through the computational model. Furthermore, from the viewpoint of partner robots, the model of emotions may help us to build robots that can have empathy for humans. To understand and sympathize with people's feelings, the robots need to have their own emotions. This may allow robots to be accepted in human society. The proposed model is implemented using deep neural networks consisting of three modules, which interact with each other. I simulated the behavior of the model with tasks that mimic mother-child interaction.