The demand for virtual private networks (VPNs) that provide confidentiality, integrity, and authenticity of communications is growing every year. IPsec is one of the oldest and most widely used VPN protocols, implemented between the internet protocol (IP) layer and the data link layer of the Linux kernel. This implementation method, known as bump-in-the-stack, has the advantage of being able to transparently apply IPsec to traffic without changing the application. However, its throughput efficiency (Gbps/core) is worse than regular Linux communication. Therefore, we chose the bump-in-the-wire (BITW) architecture, which handles IPsec in hardware separate from the host. Our proposed BITW architecture consists of inline cryptographic accelerators implemented in field-programmable gate arrays and a programmable switch that connects multiple such accelerators. A VPN gateway implemented with our architecture is transparent and improves the throughput efficiency by 3.51 times and power efficiency by 3.40 times over a VPN gateway implemented in the Linux kernel. It also demonstrates excellent scalability, and has been confirmed to scale to a maximum of 386.24 Gbps per tunnel, exceeding state-of-the-art technology in maximum throughput and efficiency per tunnel. In multiple-tunnels use cases, the proposed architecture improves the energy efficiency by 2.49 times.

VTA-NIC Chip ArchitectureWe aim to achieve DL inference serving (DLIS) without CPU interference.We integrate hardware data paths as a NIC (Network Interface Card), a REST API parser/deparser and multiple VTAs (Versatile Tensor Accelerators).ConfigurationVTA-NICProcess node16 nm FinFET @Xilinx FPGANumber of Cores8 VTA CoresCore Frequency213 MHzMACs per core169Memory Throughput19.2GB/s (DDR4-2400)Number precisionINT8PerformancePower EfficiencyThe DLIS power efficiency of VTA-NIC is 6.1x better than that of GPU (Nvidia V100).Tail LatencyAt high load, the tail latency of heterogeneous systems unexpectedly increases. With our chip, the tail latency is predictable since it is proportional to the load.BackgroundRecently, web applications are often built on microservices.DL Inference Serving (DLIS) is one of those microservices1.DLIS is provisioned with a special accelerator instance2.The microservices/instances are loosely coupled via APIs.BackgroundAccelerator instances risk inefficient data movement.1.Moving data via host processors decreases the accelerator's utilization.3a.In our preliminary experiments, half of the DLIS latency was caused by moving data.2.Under high-load conditions, the interference of host processors degrades DLIS tail latency by up to 100 times4.a.In the real cloud, 9% of light DLIS tasks suffer server tail latency, and half of the serving time is waiting time.5Model: ResNet-18@TensorRTPrecision: INT8System: Triton Inference ServerAccelerator: Nvidia V100

Studies have shown that a person's current body state can affect his thinking (embodied cognition). We tested how value judgments can be changed by this. The results of our experiment demonstrated that participants tended to discount a future reward less when looking up than when looking down. Moreover, we found that the β parameter, which represents the value of immediate rewards relative to delayed rewards received at another point in time, significantly differed between the two conditions; whereas the δ parameter, the discount rate in the standard exponential formula, did not show a significant difference. Using functional magnetic resonance imaging, McClure et al. (2004) showed that β is mediated by the lower level, automatic processes of the limbic structures; δ is mediated by the lateral prefrontal cortex, supporting higher cognitive functions. Combining the above two results, we can conclude that the embodied cognition in our experiment was mainly produced by the lower level brain processes. We believe that the knowing that discount factor β can be controlled by posture can be applied when designing robot behavior, such as encouraging a diet, trying to sell insurance, or offering customers a card loan.

It is important that participants form a positive impression when they meet with a new robot for the first time. Therefore, we tried to change how participants evaluate a new robot by changing the surrounding environment in the experiment. Embodied Cognition is a theory that our cognition for a target is strongly influenced by physical stimulation of our own bodies. Our research investigated a hypothesis that a soft chair can make favorable impressions of a robot. We performed an experiment using a hard plastic chair and a soft cushioned chair. Although the results did not support the hypothesis, they suggested that the effect of embodied cognition can be different between males and females.

Similar to relationships between humans, a person desiring to form a good relationship with a robot needs to be able to empathize with it. However, the specific kinds of human-robot interactions that would arouse and enhance empathy for the robot in the user's mind have not yet been clarified. In addition, the human behavioral traits that may be regarded as indices of empathy have not been investigated extensively. In an attempt to address these two issues, a preliminary experiment on empathy in human-robot interaction is conducted. The results suggest that the actions of naming or comforting a robot could contribute to enhancing its user's empathy and that eye fixation could be used as an index of empathy even when the use of a subjective index is inconclusive.

We implemented a humanoid robot, ASKA, in our university reception desk for the computerized university guidance. ASKA can recognize a user's question utterance, and answer the user's question by its text-to-speech voice, hand gesture and head movement. This paper describes the speech related parts of ASKA. ASKA can deal with a wide task domain of 20k large vocabulary using a word trigram model and an elaborated speaker-independent acoustic model. ASKA can also make a response with keyword and key-phrase detection in the N-best speech recognition results. The word recognition rate for the reception task is 90.9%, and the rate for the out-of-domain task is 78.9%. The correct response rate for the reception task is 61.7%. Users can enjoy their question-answering with ASKA.