Searcharxiv⌕ Search

arXiv subjects

Pratibha Verma

Publications and source records attributed to Pratibha Verma.

6 recordsLinked to original sources

SkipLane: Write-Once, Conflict-Free Accumulation for Zero-Skipping CNN Engines on FPGAs

Skipping zero activations saves multiplications in CNN accelerators, but it complicates accumulation: the products in a compressed input window belong to different output columns and must be merged. In our earlier engine, this merging was done by read-modify-write (RMW) of a row buffer, which made sparse inputs slower than dense ones even at 87% sparsity. Existing designs either merge kernel rows through RMW or reduce bank conflicts with extra hardware, without eliminating them. SkipLane makes the entire accumulation path write-once. Each kernel row is processed in its own lane, which holds partial sums in three registers and retires up to three finished columns per cycle into memory banks that cannot conflict. Lanes and input channels are summed only at readout, so no partial sum is ever read back. Multipliers left idle by zero weights compute a second filter in the same pass. In RTL simulation, the engine stays within 0.5% of its frontend bound up to 70% sparsity and is 1.10-1.26$\times$ faster than an SCNN-style scatter accumulator with 18 banks and ideal RMW at 50-80% sparsity. Six engines meet 140 MHz on a PYNQ-Z2 using 21.1k LUTs and run LeNet-5 bit-exact on the board. For conv1, the engine needs 1.94$\times$ fewer cycles than a dense engine (1.26$\times$ faster end to end); for conv2, whose rows are only 13 pixels wide, it reaches 0.87$\times$ the dense speed.

cs.AR↗

A flexible language model-assisted electronic design automation framework

Large language models (LLMs) are transforming electronic design automation (EDA) by enhancing design stages such as schematic design, simulation, netlist synthesis, and place-and-route. Existing methods primarily focus these optimisations within isolated open-source EDA tools and often lack the flexibility to handle multiple domains, such as analogue, digital, and radio-frequency design. In contrast, modern systems require to interface with commercial EDA environments, adhere to tool-specific operation rules, and incorporate feedback from design outcomes while supporting diverse design flows. We propose a versatile framework that uses LLMs to generate files compatible with commercial EDA tools and optimise designs using power-performance-area reports. This is accomplished by guiding the LLMs with tool constraints and feedback from design outputs to meet tool requirements and user specifications. The framework is used to evaluate case studies on operational transconductance amplifiers, microstrip patch antennas, and FPGA circuits.

eess.SY↗

AgenticSizing: A Large Language Model-based Multi-Agent Framework for Analog Circuit Sizing

Analog circuit sizing remains a challenging and time-consuming task due to the large design space, strong performance trade-offs, and increasing circuit complexity in scaled technologies. Although recent large language model (LLM)-based methods show promise in improving sample efficiency and interpretability, existing approaches often lack explicit circuit-topology understanding and are mainly evaluated on relatively simple analog building blocks. This paper presents a multi-agent LLM-based framework for complex analog circuit sizing. The proposed framework first analyzes the circuit topology and decomposes the netlist into functional blocks and substructures. It also extracts lightweight design knowledge for reuse. Based on the extracted topology and knowledge, a planner coordinates multiple role-specialized sizing agents to update design variables and achieve global performance specifications. This workflow mimics the collaborative process of an expert analog design team and provides a structured, interpretable, and simulation-driven optimization procedure. The framework was validated on eight circuits, with the largest design containing up to 55 transistors and 60 sizing variables. Notably, for the LDO benchmark, the proposed method achieved a 60\% success rate with an average of 83 iterations, where classical optimizers failed to find feasible solutions. Further, ablation studies demonstrate that topology understanding, design-knowledge infusion, and agent specialization provide complementary benefits. The source code is available to support reproducibility.

cs.AI↗

Global Existence and Mass Decay Analysis of solutions to the discrete Redner-Ben-Avraham-Kahng coagulation model

The Redner-Ben-Avraham-Kahng (RBK) coagulation model provide a fundamental framework for modeling the aggregation of particles in various physical and biological systems. In this paper, we investigate the global existence of solutions to the discrete version of RBK coagulation equations, encompassing a wide range of coagulation kernels. Furthermore, we demonstrate that the mass decays with time and will eventually reach to zero even if the original mass is not finite. A small size particle is produced in the RBK coagulation model when two large particles collide with a slight size difference. Dealing with the role played by large size particles in the creation of small size particles was the major challenge of this research.

math.CA↗

Convergence of the discrete Redner-Ben-Avraham-Kahng coagulation equation

This article looks at the relationship between the discrete and the continuous Redner-Ben-Avraham-Kahng (RBK) coagulation models. On the basis of a priori estimation, a weak stability principle and the weak compactness in $L_1$ for the continuous RBK model is shown. By employing a sequence of discrete models to approximate the continuous one, we show that how discrete model eventually converges to the the modified continuous one using the stability principle.

math.FA↗

Existence of solutions to the continuous RednerBen-AvrahamKahng coagulation equation

We take into account a coagulation model that simulates a distinct kind of dynamics. In this model, two particles collide to produce a single particle, but the resulting particle decreases in size, allowing each particle to be fully identified by its size. It is demonstrated that the corresponding evolving integral partial differential equation has solutions for product-type coagulation kernels i.e. $0 \le \mathfrak{K}(\varrho, ς)= \mathfrak{K}(ς, \varrho)=r(ς) r(\varrho)+α(ς, \varrho), (ς, \varrho)\in \mathbb{R}_+^2$ and $\sup_{ς\in [0,R]} \frac{\mathfrak{K}(ς, \varrho)}{\varrho} \to 0 \ \mbox{as} \ \varrho \to \infty$.

math.CA↗