Suman Datta

Md Rahatul Islam Udoy, Dharanidhar Dang, Wantong Li et al.

This paper presents ThermoPix, a CMOS-compatible electronic-photonic architecture for high-spatial-resolution temperature sensing. The proposed system converts temperature-induced wavelength shifts in a photonic interferometric sensor into timing information that can be processed by CMOS circuitry. We use a valley photonic crystal Mach-Zehnder interferometer (VPCMZI) as the sensing element, whose temperature-dependent spectral response is detected using an integrated waveguide photodetector and translated into a time-varying photocurrent. A CMOS readout circuit employing a phase-transition-material device performs threshold detection and generates a timing signal corresponding to the temperature-dependent crossing event. Circuit-level simulations demonstrate a temperature sensitivity of 3.15 ns/K, a row readout time of 2 us, and a sensing power-delay product (PDP) of 0.152 fJ. The required optical power per photonic cell is 150 nW, enabling energy-efficient array operation without requiring cooling or special environmental arrangements. We also present alternative photonic layer architectures for optical power distribution across the array. In one approach, we use different tap ratios along the row, while the other uses identical tap ratios with bidirectional excitation. The resulting average photonic cell pitches are 23.26 um and 38.52 um, respectively. The proposed ThermoPix architecture therefore provides a scalable platform for integrated temperature sensing arrays that combine photonic sensing elements with CMOS-compatible timing-based readout.

Gyujun Jeong, Sungwon Cho, Minji Shon et al.

Ferroelectric field-effect transistors (FeFET)-based vertical NAND (Fe-VNAND) has emerged as a promising candidate to overcome z-scaling limitations with lower programming voltages. However, the data retention of 3D Fe-VNAND is hindered by the complex interaction between charge detrapping and ferroelectric depolarization. Developing optimized device designs requires exploring an extensive parameter space, but the high computational cost of conventional Technology Computer-Aided Design (TCAD) tools makes such wide-scale optimization impractical. To overcome these simulation barriers, we present a Physics-Informed Neural Operator (PINO)-based AI surrogate model designed for high-efficiency prediction of threshold voltage (Vth) shifts and retention behavior. By embedding fundamental physical principles into the learning architecture, our PINO framework achieves a speedup exceeding 10000x compared to TCAD while maintaining physical accuracy. This study demonstrates the model's effectiveness on a single FeFET configuration, serving as a pathway toward modeling the retention loss mechanisms.

Kiseok Lee, Sungwon Cho, Seongkwang Lim et al.

3D DRAM has emerged as a promising approach for continued density scaling, but its viability is limited by routing and hybrid bonding constraints to periphery, which may degrade sensing margin, latency, and array efficiency. With device characteristics and array parasitics extracted from TCAD, SPICE simulations are performed with peri logic in a CMOS-Bonded-Array (CBA). The analysis shows that the bitline strap architecture with amorphous oxide semiconductor (AOS) selectors is essential to manage routing congestion and parasitics. The optimized design achieves a bit density of 2.6 Gb/mm^2 (137 layers with Si access transistors or 87 layers with AOS), representing ~6x density scaling over D1b 2D DRAM. The design further demonstrates a nominal row cycle time (tRC) of 10.5 ns, compared to 21.3 ns in D1b, and a 60% reduction in read/write energy.

Sourav Dutta, Georgios Detorakis, Abhishek Khanna et al.

Many real-world mission-critical applications require continual online learning from noisy data and real-time decision making with a defined confidence level. Probabilistic models and stochastic neural networks can explicitly handle uncertainty in data and allow adaptive learning-on-the-fly, but their implementation in a low-power substrate remains a challenge. Here, we introduce a novel hardware fabric that implements a new class of stochastic NN called Neural-Sampling-Machine that exploits stochasticity in synaptic connections for approximate Bayesian inference. Harnessing the inherent non-linearities and stochasticity occurring at the atomic level in emerging materials and devices allows us to capture the synaptic stochasticity occurring at the molecular level in biological synapses. We experimentally demonstrate in-silico hybrid stochastic synapse by pairing a ferroelectric field-effect transistor -based analog weight cell with a two-terminal stochastic selector element. Such a stochastic synapse can be integrated within the well-established crossbar array architecture for compute-in-memory. We experimentally show that the inherent stochastic switching of the selector element between the insulator and metallic state introduces a multiplicative stochastic noise within the synapses of NSM that samples the conductance states of the FeFET, both during learning and inference. We perform network-level simulations to highlight the salient automatic weight normalization feature introduced by the stochastic synapses of the NSM that paves the way for continual online learning without any offline Batch Normalization. We also showcase the Bayesian inferencing capability introduced by the stochastic synapse during inference mode, thus accounting for uncertainty in data. We report 98.25%accuracy on standard image classification task as well as estimation of data uncertainty in rotated samples.

Georgios Detorakis, Sourav Dutta, Abhishek Khanna et al.

Multiplicative stochasticity such as Dropout improves the robustness and generalizability of deep neural networks. Here, we further demonstrate that always-on multiplicative stochasticity combined with simple threshold neurons are sufficient operations for deep neural networks. We call such models Neural Sampling Machines (NSM). We find that the probability of activation of the NSM exhibits a self-normalizing property that mirrors Weight Normalization, a previously studied mechanism that fulfills many of the features of Batch Normalization in an online fashion. The normalization of activities during training speeds up convergence by preventing internal covariate shift caused by changes in the input distribution. The always-on stochasticity of the NSM confers the following advantages: the network is identical in the inference and learning phases, making the NSM suitable for online learning, it can exploit stochasticity inherent to a physical substrate such as analog non-volatile memories for in-memory computing, and it is suitable for Monte Carlo sampling, while requiring almost exclusively addition and comparison operations. We demonstrate NSMs on standard classification benchmarks (MNIST and CIFAR) and event-based classification benchmarks (N-MNIST and DVS Gestures). Our results show that NSMs perform comparably or better than conventional artificial neural networks with the same architecture.

Insik Yoon, Matthew Jerry, Suman Datta et al.

In this letter, we quantify the impact of device limitations on the classification accuracy of an artificial neural network, where the synaptic weights are implemented in a Ferroelectric FET (FeFET) based in-memory processing architecture. We explore a design-space consisting of the resolution of the analog-to-digital converter, number of bits per FeFET cell, and the neural network depth. We show how the system architecture, training models and overparametrization can address some of the device limitations.

Abhisek Das, Satanik Panda, Suman Datta et al.

In this paper authors are going to present a Markov Decision Process (MDP) based algorithm in Industrial Internet of Things (IIoT) as a safety compliance layer for human in loop system. Though some industries are moving towards Industry 4.0 and attempting to automate the systems as much as possible by using robots, still human in loop systems are very common in developing countries like India. When ever there is a need for human machine interaction, there is a scope of health hazard. In this work we have developed a system for one such industry using MDP. The proposed algorithm used in this system learned the probability of state transition from experience as well as the system is adaptable to new changes by incorporating the concept of transfer learning. The system was evaluated on the data set obtained from 39 sensors connected to a computer numerically controlled (CNC) machine pushing data every second in a 24x7 scenario. The state changes are typically instructed by a human which subsequently lead to some intentional or unintentional mistakes and errors. The proposed system raises an alarm for the operator to warn which he may or may not overlook depending on his own perception about the present condition of the system. Repeated ignorance of the operator for a particular state transition warning guides the system to retrain the model. We observed 95.61% alarms raised by the said system are taken care of by the operator. 3.2% alarms are coming from the changes in the system which in turn used to retrain the model and 1.19% alarms are false alarms. We could not compute the error coming from the mistake performed by the human operator as there is no ground truth available for that.

Abhinav Parihar, Matthew Jerry, Suman Datta et al.

Artificial neural networks can harness stochasticity in multiple ways to enable a vast class of computationally powerful models. Electronic implementation of such stochastic networks is currently limited to addition of algorithmic noise to digital machines which is inherently inefficient; albeit recent efforts to harness physical noise in devices for stochasticity have shown promise. To succeed in fabricating electronic neuromorphic networks we need experimental evidence of devices with measurable and controllable stochasticity which is complemented with the development of reliable statistical models of such observed stochasticity. Current research literature has sparse evidence of the former and a complete lack of the latter. This motivates the current article where we demonstrate a stochastic neuron using an insulator-metal-transition (IMT) device, based on electrically induced phase-transition, in series with a tunable resistance. We show that an IMT neuron has dynamics similar to a piecewise linear FitzHugh-Nagumo (FHN) neuron and incorporates all characteristics of a spiking neuron in the device phenomena. We experimentally demonstrate spontaneous stochastic spiking along with electrically controllable firing probabilities using Vanadium Dioxide (VO$_2$) based IMT neurons which show a sigmoid-like transfer function. The stochastic spiking is explained by two noise sources - thermal noise and threshold fluctuations, which act as precursors of bifurcation. As such, the IMT neuron is modeled as an Ornstein-Uhlenbeck (OU) process with a fluctuating boundary resulting in transfer curves that closely match experiments. As one of the first comprehensive studies of a stochastic neuron hardware and its statistical properties, this article would enable efficient implementation of a large class of neuro-mimetic networks and algorithms.