arXiv:2606. 13179v1 Announce Type: cross Abstract: In recent years, driven by the computational demands of data-intensive applications such as artificial intelligence and scientific computing, analog computing has gained renewed interest.
In recent years, driven by the computational demands of data-intensive applications such as artificial intelligence and scientific computing, analog computing has gained renewed interest. Given the diversity of computational tasks and recent advancements in analog CMOS circuits and resistive memory technologies, we refer to the evolving landscape as modern analog computing.
The paper introduces a fully analog readout architecture for vector-matrix multiplication in spiking neural networks that uses voltage-to-time conversion instead of conventional current-mode circuits. By sensing the column voltage directly, the design eliminates the need for current summing and scaling circuitry, thereby reducing area and power consumption. Post-layout simulations of a 10×1 SNN in 130 nm CMOS and tests on a 64×10 digit‑classification network demonstrate the feasibility and efficiency of the proposed approach.
arXiv:2609.36584v1 Announce Type: new Abstract: Analog in-memory computing (AIMC) offers an alternative for model training by executing matrix operations directly where weights are stored. However, s...
arXiv:2607. 15123v1 Announce Type: cross Abstract: Recent FPGAs have improved deep learning (DL) inference efficiency through dedicated tensor blocks and in-BRAM computation.
arXiv:2606. 02781v1 Announce Type: cross Abstract: Deep neural networks (DNNs) have achieved state-of-the-art performance across diverse domains.
arXiv:2605. 11855v2 Announce Type: replace-cross Abstract: Sequence learning is dominated by Transformers and parallelizable recurrent neural networks (RNNs) such as state-space models, yet learning long-term dependencies remains challenging, and state-of-the-art designs trade power consumption for performance.
arXiv:2603. 13042v2 Announce Type: replace Abstract: Digital Compute-in-Memory (DCiM) accelerates neural networks by reducing data movement.
Posted by Manish Gupta, Staff Software Engineer, Google Research AI-driven technologies are weaving themselves into the fabric of our daily routines, with the potential to enhance our access to knowledge and boost our overall productivity. The backbone of these applications lies in large language models (LLMs).
arXiv:2507.19290v2 Announce Type: replace-cross Abstract: We study the problem of learning a structured approximation (low-rank, sparse, banded, etc.) to an unknown matrix $A$ given access to matrix-...
arXiv:2604. 26979v2 Announce Type: replace-cross Abstract: In-memory computing (IMC) is a paradigm that enables neural network inference by computing analog matrix-vector multiplications (MVM) directly in memory crossbar arrays, with the potential for energy efficiency gains over conventional von Neumann architectures.
arXiv:2609.37525v1 Announce Type: new Abstract: Physical neural networks and analog in-memory computing could reduce the energy cost of neural network training. Realizing this potential, however, req...