Opinion: Progress in deep learning research will come from the convergence of engineering and neuroscience.

As anyone who has purchased jewelry can attest, platinum is expensive. That’s tough for consumers but also a serious hurdle for a promising source of electricity for vehicles: the hydrogen fuel cell, which relies on platinum.
Now a research team led by Bruce E. Koel, a professor of biological and chemical engineering at Princeton University, has opened a door to finding far cheaper alternatives. In a paper published April 4 in the journal Nature Communications, the researchers reported that a chemical compound based on hafnium worked about 60 percent as effectively as platinum-related materials but at about one-fifth the cost.
“We hope to find something that is more abundant and cheaper to catalyze reactions,” said Xiaofang Yang, principal scientist at HiT Nano Inc. and visiting collaborator at Princeton who is working with Koel on the project.
Neuromorphic systems carry out robust and efficient neural computation using hardware implementations that operate in physical time. Typically they are event- or data-driven, they employ low-power, massively parallel hybrid analog/digital VLSI circuits, and they operate using the same physics of computation used by the nervous system. Although there are several forums for presenting research achievements in neuromorphic engineering, none are exclusively dedicated to this increasingly large research community. Either because they are dedicated to single disciplines, such as electrical engineering or computer science, or because they serve research communities which focus on analogous areas (such as biomedical engineering or computational neuroscience), but with fundamentally different goals and objectives. The mission of Neuromorphic Engineering is to provide a publication medium dedicated exclusively and specifically to this field. Topics covered by this publication include: Analog and hybrid analog/digital electronic circuits for implementing neural processes, such as conductances, neurons, synapses, plasticity mechanisms, photoreceptors, cochleae, etc. Neuromorphic circuits and systems for implementing real-time event-based neural processing architectures. Hardware models of neural and sensorimotor processing systems, such as selective attention systems, coordinate transformation systems, auditory and/or visual processing systems, sensory fusion systems, etc. Implementations of neural computational systems found in insects, birds, mammals, etc. Embedded neuromorphic systems, including actuated or robotic platforms which process sensory signals and interact with the environment using event-based sensors and circuits. To ensure high quality and state-of-the-art material, publications should demonstrate experimental results, using physical implementations of neuromorphic systems, and possibly show the links between the artificial system and the neural/biological one they model.
The launch was expected to encounter many technical and engineering challenges, including simplified procedures for pre-launch testing, the rocking motion of the ship and heat dissipation in a confined space.
China has become the first nation to fully own and operate a floating launch platform for its space missions.
Atomic BECs were first achieved in 1995. Although it has become easier to realize atomic BECs since their discovery, they still require very low temperatures for operation. For most purposes, this is too expensive and impractical. Alternatively, negatively charged quatrons are quasi-particles composed of a hole and three electrons which form a stable BEC when coupled to light in triple quantum layer structures in semiconductor microcavities. This allows for both the greater experimental control found in quantum optics, and the benefits of matter wave systems, such as superconductivity and coherence. Moreover, due to the extremely small effective mass of the quasi-particles, quatrons can be used to achieve superconducting BECs at room temperature.
The Create the Future Design Contest was launched in 2002 by the publishers of NASA Tech Briefs magazine to help stimulate and reward engineering innovation. The annual event has attracted more than 8,000 product design ideas from engineers, entrepreneurs, and students worldwide.
Researchers, from biochemists to material scientists, have long relied on the rich variety of organic molecules to solve pressing challenges. Some molecules may be useful in treating diseases, others for lighting our digital displays, still others for pigments, paints, and plastics. The unique properties of each molecule are determined by its structure—that is, by the connectivity of its constituent atoms. Once a promising structure is identified, there remains the difficult task of making the targeted molecule through a sequence of chemical reactions. But which ones?
Organic chemists generally work backwards from the target molecule to the starting materials using a process called retrosynthetic analysis. During this process, the chemist faces a series of complex and inter-related decisions. For instance, of the tens of thousands of different chemical reactions, which one should you choose to create the target molecule? Once that decision is made, you may find yourself with multiple reactant molecules needed for the reaction. If these molecules are not available to purchase, then how do you select the appropriate reactions to produce them? Intelligently choosing what to do at each step of this process is critical in navigating the huge number of possible paths.
Researchers at Columbia Engineering have developed a new technique based on reinforcement learning that trains a neural network model to correctly select the “best” reaction at each step of the retrosynthetic process. This form of AI provides a framework for researchers to design chemical syntheses that optimize user specified objectives such synthesis cost, safety, and sustainability. The new approach, published May 31 by ACS Central Science, is more successful (by ~60%) than existing strategies for solving this challenging search problem.
Thanks to a team of researchers from the University of Illinois at Urbana-Champaign and the University of Massachusetts Amherst, scientists are able to read patterns on long chains of molecules to understand and predict behavior of disordered strands of proteins and polymers. The results could, among other things, pave the way to develop new materials from synthetic polymers.
The lab of Charles Sing, assistant professor of chemical and biomolecular engineering at Illinois, provided the theory behind the discovery, which was then verified through experiments conducted in the lab of Sarah Perry, assistant professor of chemical engineering at UMass Amherst, and Illinois alumna. The collaborators detailed their findings in a paper titled “Designing Electrostatic Interactions via Polyelectrolyte Monomer Sequence” published in ACS (American Chemical Society) Central Science.
The colleagues set out to understand the physics behind the precise sequence of charged monomers along the chain and how it affects the polymer’s ability to create self-assembling liquid materials called complex coacervates.
. The lectures introduce our current understanding of computational intelligence and ways in which strong AI could possibly be achieved, with insights from deep learning, reinforcement learning, computational neuroscience, robotics, cognitive modeling, psychology, and more.
Ray Kurzweil is one of the world’s leading inventors, thinkers, and futurists, with a thirty-year track record of accurate predictions. Called “the restless genius” by The Wall Street Journaland “the ultimate thinking machine” by Forbes magazine, Kurzweil was selected as one of the top entrepreneurs by Inc. magazine, which described him as the “rightful heir to Thomas Edison.” PBS selected him as one of the “sixteen revolutionaries who made America.”
The ability to confine water in an enclosed compartment without directly manipulating it or using rigid containers is an attractive possibility. In a recent study, Sara Coppola and an interdisciplinary research team in the departments of Biomaterials, Intelligent systems, Industrial Production Engineering and Advanced Biomaterials for Healthcare in Italy, proposed a water-based, bottom-up approach to encase facile, short-lived water silhouettes in a custom-made adaptive suit.
In the work, they used a biocompatible polymer that could self-assemble with unprecedented degrees of freedom on the water surface to produce a thin membrane. They custom designed the polymer film as an external container of a liquid core or as a free-standing layer. The scientists characterized the physical properties and morphology of the membrane and proposed a variety of applications for the phenomenon from the nanoscale to the macroscale. The process could encapsulate cells or microorganisms successfully without harm, opening the way to a breakthrough approach applicable for organ-on-a-chip and lab-in-a-drop experiments. The results are now published in Science Advances.
The possibility of isolating, engineering and shaping materials into 2-D or 3D objects from the nanometer to the microscale via bottom-up engineering is gaining importance in materials science. Understanding the physics and chemistry of materials will allow a variety of applications in microelectronics, drug delivery, forensics, archeology and paleontology and space research. Materials scientists use a variety of technical methods for microfabrication including two-photon polymerization, soft interference lithography, replica molding and self-folding polymers to shape and isolate the material of interest. However, most materials engineering protocols require chemical and physical pretreatments to gain the desired final properties.