January 2022
Deepfakes are real and a real threat to organizations all over the world, including government organizations. Categorizing deepfakes is important and allows for separating this technology into positive and negative realms.
From autonomous vehicles to hybrid cloud and mainframe systems, Reliability, Availability and Serviceability (RAS) is fast becoming a first-class consideration alongside performance and energy-optimal execution in the design and functioning of the constituent processors. This has resulted in an increased push towards Efficient Resilience as the predominant design paradigm for these processors.
For AI-powered systems to be truly successful in practical and everyday scenarios, it is not just enough for these systems to generate optimized decisions, but they need to be capable of working and collaborating with users from all walks of life. One major requirement for developing such systems is the need to imbue them with the ability to effectively model the expectations of their users, and to be capable of explaining their decisions and the rationale behind them in intuitive terms when such expectations cannot be met.
Data explosion is facing grand computing challenges on computation power, memory capacity, and communication bandwidth. In the post-Moore’s Law era, specialization has become a new driving force to tackle these problems for performance and energy efficiency improvement.
Interested to learn more about an engineering career, from the perspective of five women with decades of experience in engineering? Bring your questions to this open Q&A panel event. Get advice. Network. This event is sponsored by EECS, however, please invite friends pursuing STEM careers who also may be interested.
Attack surfaces are defects in software and hardware that can be misused by attackers to do undesired computations (e.g., steal credentials, leak private user data, and take control over the whole system). Previous efforts on mitigating attack surfaces aim at individual security incidents. As a consequence, software systems are integrated with too many ad-hoc protections while still not becoming secure
The classic machine learning (ML) paradigm involves algorithms that learn from centralized data, possibly pooled together from multiple data sources. The computations involved may be done on a single machine or farmed out to a cluster of machines but are inherently restricted within the availability of in-house resources. As a result, such algorithms often fail to utilize the computation and communication capability of their entire ecosystem to support their ever-growing scales.
The exploding deployment of mobile and IoT devices has changed many aspects of our lives. These devices are supported by the involvement of multiple key stakeholders, such as platform vendors, device manufacturers, and third-party developers, with each having her own unique cyber threats.
There is no doubt that robots will play a crucial role in the future and need to work as a team in increasingly more complex applications. Advances in robotics have laid the hardware foundations for building large-scale multi-robot systems, such as for mobile robots and drones. But how to coordinate robots intelligently is a difficult problem.