An analog-AI chip for energy-efficient speech recognition and transcription.

Models of artificial intelligence (AI) that have billions of parameters can achieve high accuracy across a range of tasks, but they exacerbate the poor energy efficiency of conventional general-purpose processors, such as graphics processing units or central processing units. Analog in-memory computing (analog-AI) can provide better energy efficiency by performing matrix-vector multiplications in parallel on 'memory tiles'. However, analog-AI has yet to demonstrate software-equivalent (SW) accuracy on models that require many such tiles and efficient communication of neural-network activations between the tiles.

Optimised weight programming for analogue memory-based deep neural networks.

Analogue memory-based deep neural networks provide energy-efficiency and per-area throughput gains relative to state-of-the-art digital counterparts such as graphics processing units. Recent advances focus largely on hardware-aware algorithmic training and improvements to circuits, architectures, and memory devices. Optimal translation of software-trained weights into analogue hardware weights-given the plethora of complex memory non-idealities-represents an equally important task.

Toward Software-Equivalent Accuracy on Transformer-Based Deep Neural Networks With Analog Memory Devices.

Recent advances in deep learning have been driven by ever-increasing model sizes, with networks growing to millions or even billions of parameters. Such enormous models call for fast and energy-efficient hardware accelerators. We study the potential of Analog AI accelerators based on Non-Volatile Memory, in particular Phase Change Memory (PCM), for software-equivalent accurate inference of natural language processing applications.

Ethical and Methodological Considerations for Evaluating Participant Views on Alzheimer's and Dementia Research.

The urgent need to expand enrollment in Alzheimer's disease and related dementia (ADRD) research has synergized calls for an empiric science of research recruitment, yet, progress in this area is hindered by challenges to measuring views toward ADRD research. This paper reports ethical and methodological considerations identified through a prospective qualitative study investigating ADRD patient and caregiver views on research recruitment and participation surrounding acute illness. Ethical and methodological considerations were identified through a combination of memoing, collaboration with a Community Advisory Board (CAB), and analysis of interview data from ADRD patients ( = 3) and/or caregivers ( = 28).

Tailoring Research Recruitment for Acute Care Settings: Recommendations from People with Dementia and their Caregivers.

Background: There is a pressing need to increase enrollment and representation in Alzheimer's disease and related dementia (ADRD) research. Current recruitment approaches focus largely on clinic and community settings, with minimal engagement of acute care environments despite their broad use across diverse populations. The objectives of this study were to examine views, preferences, and recommendations regarding acute care-based ADRD research recruitment among persons with dementia and their caregivers.