Edward Rietman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Edward Rietman has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Hava Tova Siegelmann
Manning College of Information and Computer Sciences, University of Massachusetts Amherst
4 shared publications - Alexander Yeung
College of Computer & Information Sciences, University of Massachusetts Amherst, Amherst, MA 01003, USA
2 shared publications - Arjun Karuvally
College of Computer & Information Sciences, University of Massachusetts Amherst, Amherst, MA 01003, USA
2 shared publications - Hananel Hazan
Allen Discovery Center at Tufts University, Medford, MA 02155, USA
2 shared publications - Jack A. Tuszynski
Dipartimento di Ingegneria Meccanica e Aerospaziale, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Turin, Italy
2 shared publications - Peter DelMastro
Department of Mathematics, Virginia Tech, Blacksburg, VA 24061, USA
2 shared publications
Latest Publications
- Open Access
- 13 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 32 (2) , 1-66
Optimizing Multilinear Polynomial Representations of Boolean Functions for Neural Network Synthesis
We consider the problem of obtaining a neural network (NN) representation of a Boolean network (BN), where each Boolean network (BN) vertex is a k -input Boolean function (BF) ( k≤ K ). A neural network (NN) representation can be constructed by mapping each Boolean function (BF) to its multilinear polynomial (MP) representation and then to a two-layer sub-NN; composing sub-NNs completes the representation. In this article, we show that structural properties (number of neurons/connections, neuron in-degree) of such a NN representation can be efficiently optimized while maintaining functional equivalence with the original BN ( lossless optimization). Our approach establishes a correspondence between multilinear polynomials (MPs) and their two-layer-NN representation, enabling sub-NN optimization by minimizing MPs under structural criteria. To accelerate optimization, we utilize criterion invariance under Negation-Permutation-Negation (NPN) transformations. This allows us to classify BFs into equivalence classes, optimize canonical representatives, and then derive optimal representations for remaining class members via Negation-Permutation-Negation (NPN) transformations. Since optimizing sub-NNs may affect the composed NN depth, we propose an algorithm for lossless optimization under depth constraints. Experimentally, across four control-oriented and cryptographic digital circuits, we achieve reductions of up to 59% in neurons and 69% in connections relative to non-optimized NNs. Under minimum-depth constraints, reductions of up to 23% in neurons and 50% in connections are obtained. We demonstrate minimization of mean and maximum in-degree, and that intermediate-depth solutions recover most of the size reduction. Our NPN-based acceleration reduces MP optimization time by up to 10.2× compared to a baseline caching solutions per BF. Overall, the proposed methods facilitate high-throughput BN simulation and neurosymbolic matrix-vector-multiplication architectures.
- 13 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
- Vol. 32 (2) , 1-66
Open AccessOptimizing Multilinear Polynomial Representations of Boolean Functions for Neural Network Synthesis
We consider the problem of obtaining a neural network (NN) representation of a Boolean network (BN), where each Boolean network (BN) vertex is a k -input Boolean function (BF) ( k≤ K ). A neural network (NN) representation can be constructed by mapping each Boolean function (BF) to its multilinear polynomial (MP) representation and then to a two-layer sub-NN; composing sub-NNs completes the representation. In this article, we show that structural properties (number of neurons/connections, neuron in-degree) of such a NN representation can be efficiently optimized while maintaining functional equivalence with the original BN ( lossless optimization). Our approach establishes a correspondence between multilinear polynomials (MPs) and their two-layer-NN representation, enabling sub-NN optimization by minimizing MPs under structural criteria. To accelerate optimization, we utilize criterion invariance under Negation-Permutation-Negation (NPN) transformations. This allows us to classify BFs into equivalence classes, optimize canonical representatives, and then derive optimal representations for remaining class members via Negation-Permutation-Negation (NPN) transformations. Since optimizing sub-NNs may affect the composed NN depth, we propose an algorithm for lossless optimization under depth constraints. Experimentally, across four control-oriented and cryptographic digital circuits, we achieve reductions of up to 59% in neurons and 69% in connections relative to non-optimized NNs. Under minimum-depth constraints, reductions of up to 23% in neurons and 50% in connections are obtained. We demonstrate minimization of mean and maximum in-degree, and that intermediate-depth solutions recover most of the size reduction. Our NPN-based acceleration reduces MP optimization time by up to 10.2× compared to a baseline caching solutions per BF. Overall, the proposed methods facilitate high-throughput BN simulation and neurosymbolic matrix-vector-multiplication architectures.
- Open Access
Reservoir Computation with Networks of Differentiating Neuron Ring Oscillators
Reservoir computing is an approach to machine learning that leverages the dynamics of a complex system alongside a simple, often linear, machine learning model for a designated task. While many efforts have previously focused their attention on integrating neurons, which produce an output in response to large, sustained inputs, we focus on using differentiating neurons, which produce an output in response to large changes in input. Here, we introduce a small-world graph built from rings of differentiating neurons as a Reservoir Computing substrate. We find the coupling strength and network topology that enable these small-world networks to function as an effective reservoir. The dynamics of differentiating neurons naturally give rise to oscillatory dynamics when arranged in rings, where we study their computational use in the Reservoir Computing setting. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing Reservoir Computing approaches. Beyond accuracy, we conduct systematic analysis of our reservoir’s internal dynamics using three complementary complexity measures that quantify neuronal activity balance, input dependence, and effective dimensionality. Our analysis reveals that optimal performance emerges when the reservoir operates with intermediate levels of neural entropy and input sensitivity, consistent with the edge-of-chaos hypothesis, where the system balances stability and responsiveness. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications.
Open AccessReservoir Computation with Networks of Differentiating Neuron Ring Oscillators
Reservoir computing is an approach to machine learning that leverages the dynamics of a complex system alongside a simple, often linear, machine learning model for a designated task. While many efforts have previously focused their attention on integrating neurons, which produce an output in response to large, sustained inputs, we focus on using differentiating neurons, which produce an output in response to large changes in input. Here, we introduce a small-world graph built from rings of differentiating neurons as a Reservoir Computing substrate. We find the coupling strength and network topology that enable these small-world networks to function as an effective reservoir. The dynamics of differentiating neurons naturally give rise to oscillatory dynamics when arranged in rings, where we study their computational use in the Reservoir Computing setting. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing Reservoir Computing approaches. Beyond accuracy, we conduct systematic analysis of our reservoir’s internal dynamics using three complementary complexity measures that quantify neuronal activity balance, input dependence, and effective dimensionality. Our analysis reveals that optimal performance emerges when the reservoir operates with intermediate levels of neural entropy and input sensitivity, consistent with the edge-of-chaos hypothesis, where the system balances stability and responsiveness. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications.
- Open Access
Reservoir Computation with Networks of Differentiating Neuron Ring Oscillators
Reservoir Computing is a machine learning approach that uses the rich repertoire of complex system dynamics for function approximation. Current approaches to reservoir computing use a network of coupled integrating neurons that require a steady current to maintain activity. Here, we introduce a small world graph of differentiating neurons that are active only when there are changes in input as an alternative to integrating neurons as a reservoir computing substrate. We find the coupling strength and network topology that enable these small world networks to function as an effective reservoir. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing reservoir computing approaches. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications.
Open AccessReservoir Computation with Networks of Differentiating Neuron Ring Oscillators
Reservoir Computing is a machine learning approach that uses the rich repertoire of complex system dynamics for function approximation. Current approaches to reservoir computing use a network of coupled integrating neurons that require a steady current to maintain activity. Here, we introduce a small world graph of differentiating neurons that are active only when there are changes in input as an alternative to integrating neurons as a reservoir computing substrate. We find the coupling strength and network topology that enable these small world networks to function as an effective reservoir. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing reservoir computing approaches. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications.
- Open Access
- 17 March 2025
- Published by Auckland University of Technology (AUT) Library
- Vol. 1
Distributed Multi-Agent Lifelong Learning
Lifelong learning (LL) systems must adapt to changing environments by continuously updating their knowledge. Traditional LL paradigms assume new data come labeled and that agents learn independently. However, labeled data are scarce, especially in remote settings. We introduce the Peer Parallel Lifelong Learning (PEEPLL) framework for distributed Multi-Agent LL, where agents actively seek peer assistance when needed. Unlike classical distributed AI, where communication scales poorly, PEEPLL agents reduce communication as their knowledge evolves. To improve resilience to low-quality peer responses, we propose (a) TRUE confidence score, a compute-efficient application of Variational Autoencoder without input reconstruction, (b) REFINE algorithm to selectively accept peer responses, (c) DYMEM, Dynamic Memory-Update for storing necessary information. Each of the three contributions improves upon their baselines. In our solution to PEEPLL, agents outperform traditional LL agents, even when the latter has environmental supervision available, marking a substantial step towards scalable lifelong learning at the edge.
- 17 March 2025
- Published by Auckland University of Technology (AUT) Library
- Vol. 1
Open AccessDistributed Multi-Agent Lifelong Learning
Lifelong learning (LL) systems must adapt to changing environments by continuously updating their knowledge. Traditional LL paradigms assume new data come labeled and that agents learn independently. However, labeled data are scarce, especially in remote settings. We introduce the Peer Parallel Lifelong Learning (PEEPLL) framework for distributed Multi-Agent LL, where agents actively seek peer assistance when needed. Unlike classical distributed AI, where communication scales poorly, PEEPLL agents reduce communication as their knowledge evolves. To improve resilience to low-quality peer responses, we propose (a) TRUE confidence score, a compute-efficient application of Variational Autoencoder without input reconstruction, (b) REFINE algorithm to selectively accept peer responses, (c) DYMEM, Dynamic Memory-Update for storing necessary information. Each of the three contributions improves upon their baselines. In our solution to PEEPLL, agents outperform traditional LL agents, even when the latter has environmental supervision available, marking a substantial step towards scalable lifelong learning at the edge.
- Open Access
On the Dynamics of Learning Time-Aware Behavior with Recurrent Neural Networks
Recurrent Neural Networks (RNNs) have shown great success in modeling time-dependent patterns, but there is limited research on their learned representations of latent temporal features and the emergence of these representations during training. To address this gap, we use timed automata (TA) to introduce a family of supervised learning tasks modeling behavior dependent on hidden temporal variables whose complexity is directly controllable. Building upon past studies from the perspective of dynamical systems, we train RNNs to emulate temporal flipflops, a new collection of TA that emphasizes the need for time-awareness over long-term memory. We find that these RNNs learn in phases: they quickly perfect any time-independent behavior, but they initially struggle to discover the hidden time-dependent features. In the case of periodic "time-of-day" aware automata, we show that the RNNs learn to switch between periodic orbits that encode time modulo the period of the transition rules. We subsequently apply fixed point stability analysis to monitor changes in the RNN dynamics during training, and we observe that the learning phases are separated by a bifurcation from which the periodic behavior emerges. In this way, we demonstrate how dynamical systems theory can provide insights into not only the learned representations of these models, but also the dynamics of the learning process itself. We argue that this style of analysis may provide insights into the training pathologies of recurrent architectures in contexts outside of time-awareness.
Open AccessOn the Dynamics of Learning Time-Aware Behavior with Recurrent Neural Networks
Recurrent Neural Networks (RNNs) have shown great success in modeling time-dependent patterns, but there is limited research on their learned representations of latent temporal features and the emergence of these representations during training. To address this gap, we use timed automata (TA) to introduce a family of supervised learning tasks modeling behavior dependent on hidden temporal variables whose complexity is directly controllable. Building upon past studies from the perspective of dynamical systems, we train RNNs to emulate temporal flipflops, a new collection of TA that emphasizes the need for time-awareness over long-term memory. We find that these RNNs learn in phases: they quickly perfect any time-independent behavior, but they initially struggle to discover the hidden time-dependent features. In the case of periodic "time-of-day" aware automata, we show that the RNNs learn to switch between periodic orbits that encode time modulo the period of the transition rules. We subsequently apply fixed point stability analysis to monitor changes in the RNN dynamics during training, and we observe that the learning phases are separated by a bifurcation from which the periodic behavior emerges. In this way, we demonstrate how dynamical systems theory can provide insights into not only the learned representations of these models, but also the dynamics of the learning process itself. We argue that this style of analysis may provide insights into the training pathologies of recurrent architectures in contexts outside of time-awareness.
- Open Access
- 29 September 2021
- Published by PAGEPress Publications in Biomedical Science and Engineering
- Vol. 4 (s1)
How systems biology of cancer signaling pathways can inform optimized combination chemotherapy design
This talk presents an overview of the systems biology of cancer applied to specific protein-protein interaction signaling pathways. We argue that network analysis augmented with thermodynamics measures such as Gibbs free energy can lead to optimized design of drug combinations for cancer chemotherapy.
- 29 September 2021
- Published by PAGEPress Publications in Biomedical Science and Engineering
- Vol. 4 (s1)
Open AccessHow systems biology of cancer signaling pathways can inform optimized combination chemotherapy design
This talk presents an overview of the systems biology of cancer applied to specific protein-protein interaction signaling pathways. We argue that network analysis augmented with thermodynamics measures such as Gibbs free energy can lead to optimized design of drug combinations for cancer chemotherapy.
- Open Access
- 19 March 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (3) , e0226883
Personalized therapy design for systemic lupus erythematosus based on the analysis of protein-protein interaction networks
We analyzed protein expression data for Lupus patients, which have been obtained from publicly available databases. A combination of systems biology and statistical thermodynamics approaches was used to extract topological properties of the associated protein-protein interaction networks for each of the 291 patients whose samples were used to provide the molecular data. We have concluded that among the many proteins that appear to play critical roles in this pathology, most of them are either ribosomal proteins, ubiquitination pathway proteins or heat shock proteins. We propose some of the proteins identified in this study to be considered for drug targeting.
- 19 March 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (3) , e0226883
Open AccessPersonalized therapy design for systemic lupus erythematosus based on the analysis of protein-protein interaction networks
We analyzed protein expression data for Lupus patients, which have been obtained from publicly available databases. A combination of systems biology and statistical thermodynamics approaches was used to extract topological properties of the associated protein-protein interaction networks for each of the 291 patients whose samples were used to provide the molecular data. We have concluded that among the many proteins that appear to play critical roles in this pathology, most of them are either ribosomal proteins, ubiquitination pathway proteins or heat shock proteins. We propose some of the proteins identified in this study to be considered for drug targeting.
- 1 September 2017
- Published by Springer Nature
- p. 61-79
Thermodynamics and Cancer Dormancy: A Perspective
- 1 September 2017
- Published by Springer Nature
- p. 61-79
Thermodynamics and Cancer Dormancy: A Perspective
- 1 January 1998
- Published by American Vacuum Society in Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures Processing, Measurement, and Phenomena
- Vol. 16 (1) , 131-136
Preliminary empirical results suggesting the mapping of dynamicin situprocess signals to real-time wafer attributes in a plasma etch process
In many chemical engineering processes the input control parameters and the product output quality are monitored dynamically in real-time. Almost all the processing steps associated with semiconductor manufacturing are chemical processes in which the surface of crystals or thin films are being chemically modified. In situ monitoring of the wafer attributes in real-time is essentially nonexistent in modern semiconductor manufacturing. For plasma etching processes several new diagnostic techniques (e.g., full-wafer imaging interferometry and ellipsometry) provide improved endpoint observation and some provide metrics for the state of the wafer at a given time. However, the methods that do provide metrics are usually quite expensive for a manufacturing environment. We propose a method whereby simple and economic endpoint methods can indicate in real-time the state of the wafer. Our method consists of finding the algorithm to map in situ wafer-state signatures (e.g., interferometry, ellipsometry) to wafer attributes and then mapping the process signatures (e.g., reflected rf power, pressure, flow rate, OES) to wafer-state signatures. From these we then have an abstract mapping from the process signatures to the wafer attributes in real-time. In this article we suggest that a learning machine can perform the mapping between process signatures, as a function of time and wafer state signatures, as a function of time.
- 1 January 1998
- Published by American Vacuum Society in Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures Processing, Measurement, and Phenomena
- Vol. 16 (1) , 131-136
Preliminary empirical results suggesting the mapping of dynamicin situprocess signals to real-time wafer attributes in a plasma etch process
In many chemical engineering processes the input control parameters and the product output quality are monitored dynamically in real-time. Almost all the processing steps associated with semiconductor manufacturing are chemical processes in which the surface of crystals or thin films are being chemically modified. In situ monitoring of the wafer attributes in real-time is essentially nonexistent in modern semiconductor manufacturing. For plasma etching processes several new diagnostic techniques (e.g., full-wafer imaging interferometry and ellipsometry) provide improved endpoint observation and some provide metrics for the state of the wafer at a given time. However, the methods that do provide metrics are usually quite expensive for a manufacturing environment. We propose a method whereby simple and economic endpoint methods can indicate in real-time the state of the wafer. Our method consists of finding the algorithm to map in situ wafer-state signatures (e.g., interferometry, ellipsometry) to wafer attributes and then mapping the process signatures (e.g., reflected rf power, pressure, flow rate, OES) to wafer-state signatures. From these we then have an abstract mapping from the process signatures to the wafer attributes in real-time. In this article we suggest that a learning machine can perform the mapping between process signatures, as a function of time and wafer state signatures, as a function of time.