Neuromorphic systems that learn and predict from streaming inputs hold significant promise in pervasive edge computing and its applications. In this paper, a neuromorphic system that processes spatio-temporal information on the edge is proposed. Algorithmically, the system is based on HTM, or hierarchical temporal memory, that inherently offers online learning, resiliency, and fault tolerance. Architecturally, it is a full custom mixed-signal design with an underlying digital communication scheme and analog computational modules. Therefore, the proposed system features reconfigurability, real-time processing, low power consumption, and low-latency processing.
This study presents a novel biometric approach to identify operators, given only streams of their control movements within a manual control task setting. In the present task subjects control a simulated, remotely operated robotic arm, attempting to dock onto a satellite in orbit. The proposed methodology utilizes the Hierarchical Temporal Memory (HTM) algorithm to distinguish operators by their unique control behaviors.
This paper seeks to answer the question of whether a neocortex-inspired model called MCN can combine with entorhinal cortex cells for space representation to exploit additional metric data like odometry. The authors argue such a combination of bio-inspired techniques could help someday to create a biologically plausible and more robust navigation system like in animals.
In this paper, which Numenta VP Subutai Ahmad co-authored, the authors argue that despite the substantial advances deep learning has created in image and speech recognition, language translation, and game playing, the AI community should focus less on methods using nonbiological ANNs and more on the computational principles of the brain to create human-level or general AI with a performance close to humans at most cognitive tasks of interest.
In this paper, we investigate and evaluate the use of hierarchical temporal memory (HTM) for short-term prediction of traffic flows over real-world Sydney Coordinated Adaptive Traffic System data on arterial roads in the Adelaide metropolitan area in South Australia. Results are compared with those from long-short-term memory (LSTM). Extended experimentation with LSTM network configurations in both batch learning and online learning modes provide results with superior predictive performance over previous usage of LSTM and other deep learning techniques for short-term traffic flow prediction. In addition, we argue that HTM has potential as an effective tool for short term traffic flow prediction with results on par with LSTM and improvements when traffic flow distributions change.