An Chen
An Chen has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Charles Mackin
IBM Research
4 shared publications - Geoffrey William Burr
IBM Research—Almaden, San Jose, USA
4 shared publications Alexander Friz
IBM Research
3 shared publications- Andrea Fasoli
IBM Research – Almaden, San Jose, USA
3 shared publications - Hsinyu Tsai
IBM Research—Almaden, San Jose, USA
3 shared publications Pritish Narayanan
IBM Research--Almaden
3 shared publications
Latest Publications
- Open Access
- 30 October 2025
- Published by Springer Nature in Research Square
Row-by-row convolutional neural networks for Analog-AI
Analog AI implements the multiply-accumulate operations that dominate deep learning at the location of the weight data, offering orders of magnitude performance and energy improvements over conventional digital systems. However, translating these benefits to the Convolutional Neural Networks (CNN) widely used in applications such as image and speech processing is non-trivial. Significant reuse of both weights and activations, together with the need to extensively rearrange activations between layers, require micro-architectural solutions that span across weight mapping strategy, activation positioning, pipelining between stages, and data transport. In this paper, we describe a weight-stationary Analog AI micro-architecture for Convolutional Neural Networks (CNNs), called Row-By-Row- (RBR-) CNNs and its associated circuits and pipelines. We show a hardware demonstration of RBR-CNNs on a 14nm Analog AI inference chip with Phase Change Memory (PCM), achieving software-equivalent accuracy on the ML Perf Benchmark “Key Word Spotting” task using 4 RBR CNN layers. We show RBR CNNs using Analog AI can achieve 7x–15x latency improvements vs. high performance chips reported on ML Perf, while offering extremely high energy-efficiency – at least two orders of magnitude better than published low-power edge chip results, indicating strong applicability in embedded and mobile settings.
- 30 October 2025
- Published by Springer Nature in Research Square
Open AccessRow-by-row convolutional neural networks for Analog-AI
Analog AI implements the multiply-accumulate operations that dominate deep learning at the location of the weight data, offering orders of magnitude performance and energy improvements over conventional digital systems. However, translating these benefits to the Convolutional Neural Networks (CNN) widely used in applications such as image and speech processing is non-trivial. Significant reuse of both weights and activations, together with the need to extensively rearrange activations between layers, require micro-architectural solutions that span across weight mapping strategy, activation positioning, pipelining between stages, and data transport. In this paper, we describe a weight-stationary Analog AI micro-architecture for Convolutional Neural Networks (CNNs), called Row-By-Row- (RBR-) CNNs and its associated circuits and pipelines. We show a hardware demonstration of RBR-CNNs on a 14nm Analog AI inference chip with Phase Change Memory (PCM), achieving software-equivalent accuracy on the ML Perf Benchmark “Key Word Spotting” task using 4 RBR CNN layers. We show RBR CNNs using Analog AI can achieve 7x–15x latency improvements vs. high performance chips reported on ML Perf, while offering extremely high energy-efficiency – at least two orders of magnitude better than published low-power edge chip results, indicating strong applicability in embedded and mobile settings.
- Open Access
- 30 September 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-11
Demonstration of transformer-based ALBERT model on a 14nm analog AI inference chip
A Lite Bidirectional Encoder Representations from Transformers model is demonstrated on an analog inference chip fabricated at 14nm node with phase change memory. The 7.1 million unique analog weights shared across 12 layers are mapped to a single chip, accurately programmed into the conductance of 28.3 million devices, for this first analog hardware demonstration of a meaningfully large Transformer model. The implemented model achieved near iso-accuracy on the General Language Understanding Evaluation benchmark of seven tasks, despite the presence of weight-programming errors, hardware imperfections, readout noise, and error propagation. The average hardware accuracy was only 1.8% below that of the floating-point reference, with several tasks at full iso-accuracy. Careful fine-tuning of model weights using hardware-aware techniques contributes an average hardware accuracy improvement of 4.4%. Accuracy loss due to conductance drift – measured to be roughly 5% over 30 days – was reduced to less than 1% with a recalibration-based “drift compensation” technique.
- 30 September 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-11
Open AccessDemonstration of transformer-based ALBERT model on a 14nm analog AI inference chip
A Lite Bidirectional Encoder Representations from Transformers model is demonstrated on an analog inference chip fabricated at 14nm node with phase change memory. The 7.1 million unique analog weights shared across 12 layers are mapped to a single chip, accurately programmed into the conductance of 28.3 million devices, for this first analog hardware demonstration of a meaningfully large Transformer model. The implemented model achieved near iso-accuracy on the General Language Understanding Evaluation benchmark of seven tasks, despite the presence of weight-programming errors, hardware imperfections, readout noise, and error propagation. The average hardware accuracy was only 1.8% below that of the floating-point reference, with several tasks at full iso-accuracy. Careful fine-tuning of model weights using hardware-aware techniques contributes an average hardware accuracy improvement of 4.4%. Accuracy loss due to conductance drift – measured to be roughly 5% over 30 days – was reduced to less than 1% with a recalibration-based “drift compensation” technique.
- Top Cited PapersOpen Access
- 30 August 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-18
Hardware-aware training for large-scale and diverse deep learning inference workloads using in-memory computing-based accelerators
Analog in-memory computing—a promising approach for energy-efficient acceleration of deep learning workloads—computes matrix-vector multiplications but only approximately, due to nonidealities that often are non-deterministic or nonlinear. This can adversely impact the achievable inference accuracy. Here, we develop an hardware-aware retraining approach to systematically examine the accuracy of analog in-memory computing across multiple network topologies, and investigate sensitivity and robustness to a broad set of nonidealities. By introducing a realistic crossbar model, we improve significantly on earlier retraining approaches. We show that many larger-scale deep neural networks—including convnets, recurrent networks, and transformers—can in fact be successfully retrained to show iso-accuracy with the floating point implementation. Our results further suggest that nonidealities that add noise to the inputs or outputs, not the weights, have the largest impact on accuracy, and that recurrent networks are particularly robust to all nonidealities.
- 30 August 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-18
Top Cited PapersOpen AccessHardware-aware training for large-scale and diverse deep learning inference workloads using in-memory computing-based accelerators
Analog in-memory computing—a promising approach for energy-efficient acceleration of deep learning workloads—computes matrix-vector multiplications but only approximately, due to nonidealities that often are non-deterministic or nonlinear. This can adversely impact the achievable inference accuracy. Here, we develop an hardware-aware retraining approach to systematically examine the accuracy of analog in-memory computing across multiple network topologies, and investigate sensitivity and robustness to a broad set of nonidealities. By introducing a realistic crossbar model, we improve significantly on earlier retraining approaches. We show that many larger-scale deep neural networks—including convnets, recurrent networks, and transformers—can in fact be successfully retrained to show iso-accuracy with the floating point implementation. Our results further suggest that nonidealities that add noise to the inputs or outputs, not the weights, have the largest impact on accuracy, and that recurrent networks are particularly robust to all nonidealities.
- 26 May 2022
- Published by Springer Nature
- p. 43-59
Accelerating Deep Neural Networks with Phase-Change Memory Devices
- 26 May 2022
- Published by Springer Nature
- p. 43-59
Accelerating Deep Neural Networks with Phase-Change Memory Devices