The notion of developing an artificial intelligence system that works similarly to the human brain is captivating and reflective, catching the attention of our readership after being highlighted in freelance writer Kathryn Hulick’s report, “Making AI think more like your brain.” Kukick’s account of neuromorphic AI, which seeks to enhance and make agile AI an ability through brain-inspired hardware, resonated with and utilized their experience where the technology stands today and where it is headed.
Echoes of Early Computing in AI’s Resource Demands
Reader Gary Pokorny made clear how the floppy disks of his personal Apple IIe posed significant resource demands towards his personal computing. He recalled, “The first computer I used … was an Apple IIe, in which I would insert one floppy disk to load word processing instructions, then take it out and insert a blank floppy to save my work, and back and forth while writing.” His memory “makes sense of why mainstream AI requires huge resources for both memory and processing. I have a harder time grasping but am fascinated by the idea of spiking neural networks merging [memory and processing] more efficiently and more like our brains.” With this single reflection, Pokorny evokes the strikingly clear logic behind the neuromorphic principles intended to unify memory and processing.
Anticipated Neural Pruning in Neuromorphic Computing
Linda Ferrazzara‘s analysis on ‘Neuromorphic AI’ sparked the interest of the readers because she correlated the development of the human brain to the AI. “All the while, I couldn’t help thinking of how human brains develop, with an initial surpluse of neurons and connections that get gradually pared down as the brain is pruned into a more efficient configuration, from prebirth to adulthood,” noted Ferrazzara. Her comment highlights the biological reason behind the inspiration of “neuromorphic computing” which can be attributed to efficiency derived not only from innovative silicious realizations but also from processes of refinement and optimization akin to synaptic pruning (“Brain Silent Areas” 261).
The Quantum Inquiry: Synthesizing Neuromorphic and Quantum Computing
To the latest of computer science, Ferrazzara mulled the possibility of integrating quantum computers with neuromorphic systems. Quantum computers perform advanced computations using principles of Quantum physics like superposition (“Schrodinger’s Cat Paradox and Quantum Computing in Holmes” 130), where subatomic particles can exist in multiple states at once, and entanglement (mui kaoe), a strange link between particles.
To answer this fascinating problem, a computer scientist from MIT, Daniela Rus, shared her insights on the differences between the two technologies. As Rus put it, “I don’t think you can directly adapt quantum computers into neuromorphic computers, but we might be able to use neuromorphic processes to control quantum computers.” Still, she provided a suggestion about possible synergy saying, “ideas from quantum mechanics may be useful to design new chips for neuromorphic computers.”
One proposal of this complementarity comes from Prasanna Date, a computer scientist at Oak Ridge National Laboratory, who states, “quantum and neuromorphic computers could be used to perform different but complementary computations.” To illustrate his point, he explains, “a quantum computer could be used to train spiking neural network models, which get deployed on a neuromorphic computer for energy-efficient, real-time machine learning computations.” With this perspective, Date envisions powerful quantum machines being put to work testing and refining brain-inspired AI systems to ultimately be deployed in real time, optimized for the most energy-efficient performance.
Corrections and Clarifications
In other news, the publication provided corrections to two prior articles. In the February feature of “Holding back a glacier,” the first image was misattributed to Thwaites Glacier when, in fact, it was Pine Island Glacier. Moreover, in March issue’s “Have 5 years of COVID-19 readied us for what’s next?” article, a missing word was pointed out in the second paragraph’s last sentence that should have read: “Approximately 17,000 people in the United States died of COVID-19 in the last week of that year.” With these changes, the publication’s accuracy and integrity of reporting is maintained.
