RESUMEN
Despite the advancement of machine learning techniques in recent years, state-of-the-art systems lack robustness to "real world" events, where the input distributions and tasks encountered by the deployed systems will not be limited to the original training context, and systems will instead need to adapt to novel distributions and tasks while deployed. This critical gap may be addressed through the development of "Lifelong Learning" systems that are capable of (1) Continuous Learning, (2) Transfer and Adaptation, and (3) Scalability. Unfortunately, efforts to improve these capabilities are typically treated as distinct areas of research that are assessed independently, without regard to the impact of each separate capability on other aspects of the system. We instead propose a holistic approach, using a suite of metrics and an evaluation framework to assess Lifelong Learning in a principled way that is agnostic to specific domains or system techniques. Through five case studies, we show that this suite of metrics can inform the development of varied and complex Lifelong Learning systems. We highlight how the proposed suite of metrics quantifies performance trade-offs present during Lifelong Learning system development - both the widely discussed Stability-Plasticity dilemma and the newly proposed relationship between Sample Efficient and Robust Learning. Further, we make recommendations for the formulation and use of metrics to guide the continuing development of Lifelong Learning systems and assess their progress in the future.
Asunto(s)
Educación Continua , Aprendizaje AutomáticoRESUMEN
With good wayfinding, 72% of vision impaired people who normally use paratransit would prefer to use public transport services. However, good wayfindinfg services are hard to find. In the 30 years since the passing of the Americans with Disability Act (ADA) public policy has driven significant investment and advancements in accessibility in the United States. However, this has not yet led to a commensurate realization of freedom and independence for people with disabilities due to sustained psychosocial barriers. This paper describes the considerable progress being made in standards and technology for the delivery of wayfinding services to assist people in overcoming these barriers, whether their disabilities are visual, physical, or cognitive. Using an impact study of real-world trials of a novel infrastructure-free location technology for the provision of turn-by-turn instructions across indoor and wide-area environments, the authors will show how it is possible to deliver truly independent end-to-end access to public spaces and transport services. The Waymap technology uses wearable IoT devices and sensors within the smartphone to deliver an interactive audio-based augmented reality to guide and engage people as they explore the spaces around them. The paper focuses on how to scale wayfinding and travel services to be truly universal and the vital role played by nationally and internationally recognized standards with user involvement at their core.