The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
This work presents the design and analysis of a mixed-signal neuron (MS-N) for convolutional neural networks (CNN) and compares its performance with a digital neuron (Dig-N) in terms of operating frequency, power and noise. The circuit- level implementation of the MS-N in 65 nm CMOS technology exhibits 2-3 orders of magnitude better energy-efficiency over Dig-N for neuromorphic computing applications...
Provides an abstract for each of the keynote presentations and a brief professional biography of each presenter. The complete presentations were not made available for publication as part of the conference proceedings.
Device, interconnect scaling and interconnection bottleneck are among the major challenges for CMOS scaling. Furhtermore, signal integrity issues like crosstalk-leakage of charge between capacitively coupled nets among neighboring signal lines-is becoming inexorable. We propose to astutely turn this detrimental effect into an advantage by engineering the interference among signal lines. Our proposal...
We demonstrate multi-level optical weights embedded in a silicon photonic platform based on ferroelectric domain switching. Ferroelectric barium titanate integrated on silicon resonator structures is used as the memory material. By applying short voltage pulses of 100ns, we can switch fractions of the ferroelectric domains and thus change the transmission of the waveguides by more than one order of...
We propose a hybrid cryogenic memory architecture comprising of Josephson junction and Toggle MRAM. Comparison with existing cryogenic memory builds suggests that this hybrid build is a viable candidate of memory architecture for superconducting computing applications.
Coherent Ising machines are a type of optical accelerators that can solve different optimization tasks by encoding the problem in the connection matrix of the network. So far, experimental realizations have been limited to time multiplexed solutions, in which one nonlinear node is present in a feedback loop. In Hewlett Packard Labs, we investigate the implementation of a spatially multiplexed solution,...
We present progress on the construction and operation of a room- temperature quantum computer built with trapped atomic ion qubits. Based on the technological underpinnings of atomic clocks that define time, atomic qubits are standards of quantum information because they are all identical. They present a fundamentally scalable approach to quantum computation where interactions can be faithfully replicated...
Probabilistic and neural approaches, through their incorporation of nonlinearities and compression of states, enable a broader sampling of the phase space. For a broad set of complex questions that are encountered in conventional computation, this approach is very effective. In these patterns-oriented tasks a fluctuation in the size of data is akin to a thermal fluctuation. A thermodynamic view naturally...
The focus of the computing industry continues to shift towards designing and building intelligent systems that can handle and learn from large amounts of data. The availability of powerful processing hardware like GPUs and TPUs, has powered the tremendous success of many sophisticated resource intensive machine learning algorithms. However as device scaling and energy dissipation fast approach the...
Convolutional neural networks have become an essential element of spatial deep learning systems. In the prevailing architecture, the convolution operation is performed with Fast Fourier Transforms (FFT) electronically in GPUs. The parallelism of GPUs provides an efficiency over CPUs, however both approaches being electronic are bound by the speed and power limits of the interconnect delay inside the...
Optically active Spatial-Spectral (S2) materials are a unique resource for spectrally based optical memory and processing. At cryogenic temperatures, the rare-earth ions in these materials individually exhibit narrow optical resonances on the order of MHz to sub-KHz, but are inhomogeneously broadened over GHz to THz spectral bandwidths providing up to 10^7 resolvable spectral channels. The material...
The physical constraints underlying the concept of quantum circuit are considered. In particular it is shown that the point of departure for their modeling starts from the interconnection of the components into a classical network, followed by quantization of the latter, and not by the interconnection of already quantized components. The procedure is straightforward for lossless networks but cannot...
Quantum-dot cellular automata (QCA) is a paradigm for low-power, general-purpose, classical computing beyond the transistor era. In classical QCA, the elementary device is a cell, a system of quantum dots with a few mobile charges occupying some dots. Device switching is achieved by quantum mechanical tunneling between dots, and cells are interconnected locally via the electrostatic field. Logic is...
The feasibility of using commercial CMOS processes for implementing scalable cryogenic control electronics for universal quantum computers is investigated. Using a systems engineering approach, we break the system down into sub-systems and model the individual components down to transistor level. First results for area demand and power consumption indicate that even with a standard CMOS process, it...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.