Intel and Cornell University today published a joint paper demonstrating the ability of Intel’s neuromorphic chip, Loihi, to learn and recognize 10 hazardous materials from smell — even in the presence of “significant” data noise and occlusion. The coauthors say it shows how neuromorphic computing could be used to detect the precursor smells to explosives, narcotics, polymers, and more.
In the study, which was published this week in the journal Nature Machine Intelligence, the Intel- and Cornell-affiliated researchers describe “teaching” Loihi odors by configuring the circuit diagram of biological olfaction, drawing from a data set consisting of the activity of 72 chemical sensors in response to various smells. They say that their technique didn’t disrupt the chip’s memory of the scents and that it achieved “superior” recognition accuracy compared with conventional state-of-the-art methods, including a machine learning solution that required 3,000 times more training samples per class to reach the same level of classification accuracy.
Nabil Imam, a neuromorphic computing lab senior research scientist at Intel, believes the research will pave the way for neuromorphic systems that can diagnose diseases, detect weapons and explosives, find narcotics, and spot signs of smoke and carbon monoxide.
“We are developing neural algorithms on Loihi that mimic what happens in your brain when you smell something,” he said in a statement. “This work is a prime example of contemporary research at the crossroads of neuroscience and artificial intelligence and demonstrates Loihi’s potential to provide important sensing capabilities that could benefit various industries.”
Neuromorphic engineering, also known as neuromorphic computing, describes the use of circuits that mimic the nervous system’s neuro-biological architectures. Researchers at Intel, IBM, HP, MIT, Purdue, Stanford, and others hope to leverage it to develop a supercomputer a thousand times more powerful than any today.
Intel’s 14-nanometer Loihi chip has a 60-millimeter die size and contains over 2 billion transistors, 130,000 artificial neurons, and 130 million synapses, as well as three managing Lakemont cores for orchestration. Uniquely, Loihi features a programmable microcode engine for on-chip training of asynchronous spiking neural networks (SNNs), or AI models that incorporate time into their operating model such that the components of the model don’t process input data simultaneously. Intel claims this will be used for the implementation of adaptive self-modifying, event-driven, and fine-grained parallel computations “with high efficiency.”
According to Intel, Loihi processes information up to 1,000 times faster and 10,000 more efficiently than traditional processors, and it can solve certain types of optimization problems with more than three orders of magnitude gains in speed and energy efficiency. Moreover, Loihi maintains real-time performance results and uses only 30% more power when scaled up 50 times (whereas traditional hardware uses 500% more power), and it consumes roughly 100 times less energy than widely used CPU-run simultaneous location and mapping methods.
Beyond the neuromorphic computing realm, researchers at Google, the Canadian Institute for Advanced Research, the Vector Institute for Artificial Intelligence, the University of Toronto, Arizona State University, and others have investigated AI approaches to the problems of molecule identification and odor prediction. Google recently demonstrated a model that outperforms state-of-the-art approaches and the top-performing model from the DREAM Olfaction Prediction Challenge, a competition for mapping the chemical properties of odors.
Separately, IBM has developed Hypertaste, an “artificial tongue” designed to fingerprint beverages and other liquids “less fit for ingestion.”