Ian Gemp

Publications

Publications (grouped by year): Google DeepMind (2019-), UMass / Internships (2013-18), Northwestern / UTexas (2006-11).

2026 2025 2024 2023 2022 2021

Y. Bachrach, I. Gemp, M. Garnelo, J. Kramár, T. Eccles, D. Rosenbaum, T. Graepel. A Neural Network Auction For Group Decision Making Over a Continuous Space. IJCAI-21 Demonstrations Track. 2021
- Paper (PDF)
Abstract BibTex

We propose a system for conducting an auction over locations in a continuous space. It enables participants to express their preferences over possible choices of location in the space, selecting the location that maximizes the total utility of all agents. We prevent agents from tricking the system into selecting a location that improves their individual utility at the expense of others by using a pricing rule that gives agents no incentive to misreport their true preferences. The system queries participants for their utility in many random locations, then trains a neural network to approximate the preference function of each participant. The parameters of these neural network models are transmitted and processed by the auction mechanism, which composes these into differentiable models that are optimized through gradient ascent to compute the final chosen location and charged prices.
```
          @article{bachrachneural,
            title={A Neural Network Auction For Group Decision Making Over a Continuous Space},
            author={Bachrach, Yoram and Gemp, Ian and Garnelo, Marta and Kramar, Janos and Eccles, Tom and Rosenbaum, Dan and Graepel, Thore},
            journal={IJCAI-21 Demonstrations Track},
            year={2021}
          }
          
```
I. Gemp, B. McWilliams, C. Vernade, T. Graepel. EigenGame: PCA as a Nash Equilibrium. ICLR (Outstanding Paper Award). 2021
- Paper (PDF)
Video

Abstract BibTex

We present a novel view on principal component analysis (PCA) as a competitive game in which each approximate eigenvector is controlled by a player whose goal is to maximize their own utility function. We analyze the properties of this PCA game and the behavior of its gradient based updates. The resulting algorithm which combines elements from Oja's rule with a generalized Gram-Schmidt orthogonalization is naturally decentralized and hence parallelizable through message passing. We demonstrate the scalability of the algorithm with experiments on large image datasets and neural network activations. We discuss how this new view of PCA as a differentiable game can lead to further algorithmic developments and insights.
```
          @article{gemp2021eigengame,
            title={EigenGame: PCA as a Nash Equilibrium},
            author={Gemp, Ian and McWilliams, Brian and Vernade, Claire and Graepel, Thore},
            journal={ICLR},
            year={2021}
          }
          
```
R. Patel, M. Garnelo, I. Gemp, C. Dyer, Y. Bachrach. Game-theoretic Vocabulary Selection via the Shapley Value and Banzhaf Index. NAACL. 2021
- Paper (PDF)
Abstract BibTex

The input vocabulary and their learned representations are crucial to the performance of neural NLP models. Using the full vocabulary results in less explainable and more memory intensive models, with the embedding layer often constituting the majority of model parameters. It is thus common to use a smaller vocabulary to lower memory requirements and construct more interpertable models. We propose a vocabulary selection method that views words as members of a team trying to maximize the model’s performance. We apply power indices from cooperative game theory, including the Shapley value and Banzhaf index, that measure the relative importance of individual team members in accomplishing a joint task. We approximately compute these indices to identify the most influential words. Our empirical evaluation examines multiple NLP tasks, including sentence and document classification, question answering and textual entailment. We compare to baselines that select words based on frequency, TF-IDF and regression coefficients under L1 regularization, and show that this game-theoretic vocabulary selection outperforms all baselines on a range of different tasks and datasets.
```
          @inproceedings{patel2021game,
            title={Game-theoretic Vocabulary Selection via the Shapley Value and Banzhaf Index},
            author={Patel, Roma and Garnelo, Marta and Gemp, Ian and Dyer, Chris and Bachrach, Yoram},
            booktitle={Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
            pages={2789--2798},
            year={2021}
          }
          
```

2020

I. Gemp, K. R. McKee, R. Everett, E. Duéñez-Guzmán, Y. Bachrach, D. Balduzzi, A. Tacchetti. D3C: Reducing the Price of Anarchy in Multi-Agent Learning. NeurIPS Workshop: Cooperative AI. 2020
- Paper (PDF)
Abstract BibTex

Even in simple multi-agent systems, fixed incentives can lead to outcomes that are poor for the group and each individual agent. We propose a method, D3C, for online adjustment of agent incentives that reduces the loss incurred at a Nash equilibrium. Agents adjust their incentives by learning to mix their incentive with that of other agents, until a compromise is reached in a distributed fashion. We show that D3C improves outcomes for each agent and the group as a whole on several social dilemmas including a traffic network with Braess's paradox, a prisoner's dilemma, and several reinforcement learning domains.
```
          @article{gemp2020d3c,
            title={D3C: Reducing the Price of Anarchy in Multi-Agent Learning},
            author={Gemp, Ian and McKee, Kevin R. and Everett, Richard and Duéñez-Guzmán, Edgar A. and Bachrach, Yoram and Balduzzi, David and Tacchetti, Andrea},
            journal={NeurIPS Workshop: Cooperative AI},
            year={2020}
          }
          
```
T. Anthony, T. Eccles, A. Tacchetti, J. Kramár, I. Gemp, T. Hudson, N. Porcel, M. Lanctot, J. Pérolat, R. Everett, S. Singh, T. Graepel, Y. Bachrach. Learning to Play No-Press Diplomacy with Best Response Policy Iteration. NeurIPS (Spotlight). 2020
- Paper (PDF)
Abstract BibTex

Recent advances in deep reinforcement learning (RL) have led to considerable progress in many 2-player zero-sum games, such as Go, Poker and Starcraft. The purely adversarial nature of such games allows for conceptually simple and principled application of RL methods. However real-world settings are many-agent, and agent interactions are complex mixtures of common-interest and competitive aspects. We consider Diplomacy, a 7-player board game designed to accentuate dilemmas resulting from many-agent interactions. It also features a large combinatorial action space and simultaneous moves, which are challenging for RL algorithms. We propose a simple yet effective approximate best response operator, designed to handle large combinatorial action spaces and simultaneous moves. We also introduce a family of policy iteration methods that approximate fictitious play. With these methods, we successfully apply RL to Diplomacy: we show that our agents convincingly outperform the previous state-of-the-art, and game theoretic equilibrium analysis shows that the new process yields consistent improvements.
```
          @article{anthony2020learning,
            title={Learning to Play No-Press Diplomacy with Best Response Policy Iteration},
            author={Anthony, Thomas and Eccles, Tom and Tacchetti, Andrea and Kramár, János and Gemp, Ian and Hudson, Thomas C. and Porcel, Nicolas and Lanctot, Marc and Pérolat, Julien and Everett, Richard and Singh, Satinder and Graepel, Thore and Bachrach, Yoram},
            journal={NeurIPS},
            year={2020}
          }
          
```
K. R. McKee, I. Gemp, B. McWilliams, E. Duéñez-Guzmán, E. Hughes, J. Leibo. Social Diversity and Social Preferences in Mixed-Motive Reinforcement Learning. AAMAS. 2020
- Paper (PDF)
Abstract BibTex

Recent research on reinforcement learning in pure-conflict and pure-common interest games has emphasized the importance of population heterogeneity. In contrast, studies of reinforcement learning in mixed-motive games have primarily leveraged homogeneous approaches. Given the defining characteristic of mixedmotive games—the imperfect correlation of incentives between group members—we study the effect of population heterogeneity on mixed-motive reinforcement learning. We draw on interdependence theory from social psychology and imbue reinforcement learning agents with Social Value Orientation (SVO), a flexible formalization of preferences over group outcome distributions. We subsequently explore the effects of diversity in SVO on populations of reinforcement learning agents in two mixed-motive Markov games. We demonstrate that heterogeneity in SVO generates meaningful and complex behavioral variation among agents similar to that suggested by interdependence theory. Empirical results in these mixed-motive dilemmas suggest agents trained in heterogeneous populations develop particularly generalized, high-performing policies relative to those trained in homogeneous populations.
```
          @article{kmckee2020divers,
            title={Social Diversity and Social Preferences in Mixed-Motive Reinforcement Learning},
            author={McKee, Kevin and Gemp, Ian and McWilliams, Brian and Duéñez-Guzmán, Edgar and Hughes, Edward and Leibo, Joel},
            journal={AAMAS},
            year={2020}
          }
          
```
D. Balduzzi, W. Czarnecki, T. Anthony, I. Gemp, E. Hughes, J. Leibo, G. Piliouras, T. Graepel. Smooth Markets: A Basic Mechanism for Organizing Gradient-Based Learners. ICLR. 2020
- Paper (PDF)
Abstract BibTex

With the success of modern machine learning, it is becoming increasingly importantly to understand and control how learning algorithms interact. Unfortunately, negative results from game theory show there is little hope of understanding or controlling general n-player games. We therefore introduce smooth markets (SM-games), a class of n-player games with pairwise zero sum interactions. SM-games codify a common design pattern in machine learning that includes (some) GANs, adversarial training, and other recent algorithms. We show that SM-games are amenable to analysis and optimization using first-order methods.
```
          @article{balduzzi2020smoothmarkets,
            title={Smooth Markets: A Basic Mechanism for Organizing Gradient-Based Learners},
            author={Balduzzi, David and Czarnecki, Wojciech and Anthony, Thomas and Gemp, Ian and Hughes, Edward and Leibo, Joel and Piliouras, George and Graepel, Thore},
            journal={ICLR},
            year={2020}
          }
          
```

2019

I. Gemp, B. McWilliams. The Unreasonable Effectiveness of Adam on Cycles. NeurIPS Workshop: Bridging Game Theory & Deep Learning. 2019
- Paper (PDF)
Abstract BibTex

Generative adversarial networks (GANs) are state of the art generative models for images and other domains. Training GANs is difficult, although not nearly as difficult as expected given theoretical results on finding a Nash (PPAD complete) and our understanding of dynamical systems. Several new algorithms and techniques have been proposed to stabilize GAN training, but nearly all employ Adam or RMSProp. In fact, training a GAN with SGD instead of Adam often fails. Here, we aim to understand how Adam circumvents some of the difficulties associated with GAN training. To this end, we study Adam in the context of a cycle problem. The cycle problem is a canonical equilibrium problem for which naive optimization approaches, e.g., simultaneous SGD, fail. Understanding how Adam works in this context helps reveal reasons for its unexpected success.
```
          @article{gempmcwilliams2019adamcycles,
            title={The Unreasonable Effectiveness of Adam on Cycles},
            author={Gemp, Ian and McWilliams, Brian},
            journal={NeurIPS Workshop: Bridging Game Theory & Deep Learning},
            year={2019}
          }
          
```
I. Gemp, R. Nallapati, R. Ding, F. Nan, B. Xiang. Weakly Semi-Supervised Neural Topic Models. ICLR Workshop: Learning with Limited Labeled Data. 2019
- Paper (PDF)
Abstract BibTex

We consider the problem of topic modeling in a weakly semi-supervised setting. In this scenario, we assume that the user knows a priori a subset of the topics she wants the model to learn and is able to provide a few exemplar documents for those topics. In addition, while each document may typically consist of multiple topics, we do not assume that the user will identify all its topics exhaustively. Recent state-of-the-art topic models such as NVDM, referred to herein as Neural Topic Models (NTMs), fall under the variational autoencoder framework. We extend NTMs to the weakly semi-supervised setting by using informative priors in the training objective. After analyzing the effect of informative priors, we propose a simple modification of the NVDM model using a logit-normal posterior that we show achieves better alignment to user-desired topics versus other NTM models.
```
          @article{gemp2019ssntm,
            title={Weakly Semi-Supervised Neural Topic Models},
            author={Gemp, Ian and Nallapati, Ramesh and Ding, Ran and Nan, Feng and Xiang, Bing},
            journal={ICLR Workshop: Learning with Limited Labeled Data},
            year={2019}
          }
          
```

2018

I. Gemp. From Optimization to Equilibration: Understanding an Emerging Paradigm in Artificial Intelligence and Machine Learning. UMass Amherst Thesis. 2018
- Paper (PDF)
Abstract BibTex
Many existing machine learning (ML) algorithms cannot be viewed as gradient descent on some single objective. The solution trajectories taken by these algorithms naturally exhibit rotation, sometimes forming cycles, a behavior that is not expected with (full-batch) gradient descent. However, these algorithms can be viewed more generally as solving for the equilibrium of a game with possibly multiple competing objectives. Moreover, some recent ML models, specifically generative adversarial net- works (GANs) and its variants, are now explicitly formulated as equilibrium problems. Equilibrium problems present challenges beyond those encountered in optimization such as limit-cycles and chaotic attractors and are able to abstract away some of the difficulties encountered when training models like GANs.

In this thesis, I aim to advance our understanding of equilibrium problems so as to improve state-of-the-art in GANs and related domains. In the following chapters, I will present work on
1. designing a no-regret framework for solving monotone equilibrium problems in online or streaming settings (with applications to Reinforcement Learning),
2. ensuring convergence when training a GAN to fit a normal distribution to data by Crossing-the-Curl,
3. improving state-of-the-art image generation with techniques derived from theory,
4. and borrowing tools from dynamical systems theory for analyzing the complex dynamics of GAN training.
```
          @article{gemp2018thesis,
            title={From Optimization to Equilibration: Understanding an Emerging Paradigm in Artificial Intelligence and Machine Learning},
            author={Gemp, Ian},
            journal={UMass Amherst Technical Report},
            year={2018}
          }
          
```
I. Gemp, S. Mahadevan. Global Convergence to the Equilibrium of GANs using Variational Inequalities. arXiv preprint arXiv:1808.01531. 2018
- Paper (PDF)
Abstract BibTex

In optimization, the negative gradient of a function denotes the direction of steepest descent. Furthermore, traveling in any direction orthogonal to the gradient maintains the value of the function. In this work, we show that these orthogonal directions that are ignored by gradient descent can be critical in equilibrium problems. Equilibrium problems have drawn heightened attention in machine learning due to the emergence of the Generative Adversarial Network (GAN). We use the framework of Variational Inequalities to analyze popular training algorithms for a fundamental GAN variant: the Wasserstein Linear-Quadratic GAN. We show that the steepest descent direction causes divergence from the equilibrium, and guaranteed convergence to the equilibrium is achieved through following a particular orthogonal direction. We call this successful technique Crossing-the-Curl, named for its mathematical derivation as well as its intuition: identify the game's axis of rotation and move "across" space in the direction towards smaller "curling".
```
          @article{gemp2018crosscurl,
            title={Global Convergence to the Equilibrium of GANs using Variational Inequalities},
            author={Gemp, Ian and Mahadevan, Sridhar},
            journal={arXiv preprint arXiv:1808.01531},
            year={2018}
          }
          
```
B. Liu, I. Gemp, M. Ghavamzadeh, J. Liu, S. Mahadevan, M. Petrik. Proximal Gradient Temporal Difference Learning: Stable Reinforcement Learning with Polynomial Sample Complexity. Journal of Artificial Intelligence Research. 2018
- Paper (PDF)
Abstract BibTex

In this paper, we introduce proximal gradient temporal difference learning, which provides a principled way of designing and analyzing true stochastic gradient temporal difference learning algorithms. We show how gradient TD (GTD) reinforcement learning methods can be formally derived, not by starting from their original objective functions, as previously attempted, but rather from a primal-dual saddle-point objective function. We also conduct a saddle-point error analysis to obtain finite-sample bounds on their performance. Previous analyses of this class of algorithms use stochastic approximation techniques to prove asymptotic convergence, and do not provide any finite-sample analysis. We also propose an accelerated algorithm, called GTD2-MP, that uses proximal ``mirror maps'' to yield an improved convergence rate. The results of our theoretical analysis imply that the GTD family of algorithms are comparable and may indeed be preferred over existing least squares TD methods for off-policy learning, due to their linear complexity. We provide experimental results showing the improved performance of our accelerated gradient TD methods.
```
          @article{liu2018proximal,
            title={Proximal Gradient Temporal Difference Learning: Stable Reinforcement Learning with Polynomial Sample Complexity},
            author={Liu, Bo and Gemp, Ian and Ghavamzadeh, Mohammad and Liu, Ji and Mahadevan, Sridhar and Petrik, Marek},
            journal={Journal of Artificial Intelligence Research},
            volume={63},
            pages={461--494},
            year={2018}
          }
          
```

2017 2016

I. Gemp, S. Mahadevan. Online Monotone Optimization. arXiv preprint arXiv:1608.07888. 2016
- Paper (PDF)
Abstract BibTex

This paper presents a new framework for analyzing and designing no-regret algorithms for dynamic (possibly adversarial) systems. The proposed framework generalizes the popular online convex optimization framework and extends it to its natural limit allowing it to capture a notion of regret that is intuitive for more general problems such as those encountered in game theory and variational inequalities. The framework hinges on a special choice of a system-wide loss function we have developed. Using this framework, we prove that a simple update scheme provides a no-regret algorithm for monotone systems. While previous results in game theory prove individual agents can enjoy unilateral no-regret guarantees, our result proves monotonicity sufficient for guaranteeing no-regret when considering the adjustments of multiple agent strategies in parallel. Furthermore, to our knowledge, this is the first framework to provide a suitable notion of regret for variational inequalities. Most importantly, our proposed framework ensures monotonicity a sufficient condition for employing multiple online learners safely in parallel.
```
          @article{gemp2016online,
            title={Online Monotone Optimization},
            author={Gemp, Ian and Mahadevan, Sridhar},
            journal={arXiv preprint arXiv:1608.07888},
            year={2016}
          }
          
```

2015

I. Gemp. Exploring the Dynamics of Variational Inequality Games with Non-Concave Utilities. NIPS Workshop: Learning, Inference, and Control of Multi-Agent Systems. 2015
- Paper (PDF)
Abstract BibTex

Variational inequality (VI) theory has proven useful in modeling and analyzing a variety economic markets. However, in order to ensure the analysis is tractable, models are usually constrained to an unrealistic regime of concave utilities and monotone operators undermining the reliability of real-world conclusions such as the uniqueness and location of equilibria. We argue that machine learning can help address this issue. In this paper, we ignore typical monotonicity requirements and construct a generic, yet more realistic market model possessing several desirable qualities. We then borrow a tool from dynamical systems to cope with our model’s lack of theoretical guarantees. Additionally, in order to handle the large size of standard VI game formulations, we further enhance the tool to accommodate more sophisticated numerical algorithms and propose a heuristic for efficient use of generated trajectories. We illustrate these enhancements by applying the resulting pipeline in the context of cloud services.
```
          @inproceedings{gemp2015nonconcave,
            title={Exploring the Dynamics of Variational Inequality Games with Non-Concave Utilities},
            author={Gemp, Ian},
            booktitle={NIPS Workshop: Learning, Inference, and Control of Multi-Agent Systems},
            year={2015}
          }
          
```
I. Gemp, S. Mahadevan. Finding Equilibria in Large Games using Variational Inequalities. AAAI Spring Symposium. 2015
- Paper (PDF)
Abstract BibTex

In this paper, we explore an approach to computational game theory based on variational inequalities (VIs). VIs represent a comprehensive framework that provides a way to model and analyze both cooperative and non-cooperative games. Given the potentially large size of real-world games, suitable algorithms must be designed that can scale gracefully with the dimension of the problems (e.g., number of players). In this paper, we explore the effectiveness of novel Runge-Kutta methods on finding equilibrium solutions to two real-world games defined by oligopolistic economies.
```
          @inproceedings{gemp2015games,
            title={Finding Equilibria in Large Games using Variational Inequalities},
            author={Gemp, Ian and Mahadevan, Sridhar},
            booktitle={AAAI Spring Symposium},
            year={2015}
          }
          
```
I. Gemp, S. Mahadevan, B. Liu. Solving Large Sustainable Supply Chain Networks using Variational Inequalities. AAAI Workshop: Computational Sustainability. 2015
- Paper (PDF)
Abstract BibTex

In this paper, we explore a new approach to computational sustainability based on variational inequalities (VIs). Our challenge is to compute the steady state behaviors of networks of sustainable supply chains with possibly conflicting objectives. VIs provide a way to model large networks with numerous conflicting goals. Given the size of real-world networks, suitable algorithms must be selected that can scale with the dimension of the problems. In this paper, we explore the effectiveness of novel Runge-Kutta methods on finding equilibrium solutions to two real-world sustainable supply chain problems.
```
          @inproceedings{gemp2015supplychain,
            title={Solving Large Sustainable Supply Chain Networks using Variational Inequalities},
            author={Gemp, Ian and Mahadevan, Sridhar and Liu, Bo},
            booktitle={AAAI Workshop: Computational Sustainability},
            year={2015}
          }
          
```

2014

I. Gemp, S. Mahadevan. Modeling Context in Cognition using Variational Inequalities. AAAI Fall Symposium. 2014
- Paper (PDF)
Abstract BibTex

Important aspects of human cognition, like creativity and play, involve dealing with multiple divergent views of objects, goals, and plans. We argue in this paper that the current model of optimization that drives much of modern machine learning research is far too restrictive a paradigm to mathematically model the richness of human cognition. Instead, we propose a much more flexible and powerful framework of equilibration, which not only generalizes optimization, but also captures a rich variety of other problems, from game theory, complementarity problems, network equilibrium problems in economics, and equation solving. Our thesis is that creative activity involves dealing not with a single objective function, which optimization requires, but rather balancing multiple divergent and possibly contradictory goals. Such modes of cognition are better modeled using the framework of variational inequalities (VIs). We provide a brief review of this paradigm for readers unfamiliar with the underlying mathematics, and sketch out how VIs can account for creativity and play in human and animal cognition.
```
          @inproceedings{gemp2014context,
            title={Modeling Context in Cognition using Variational Inequalities},
            author={Gemp, Ian and Mahadevan, Sridhar},
            booktitle={AAAI Fall Symposium},
            year={2014}
          }
          
```
S. Mahadevan, B. Liu, P. Thomas, W. Dabney, S. Giguere, N. Jacek, I. Gemp, J. Liu. Proximal Reinforcement Learning: A New Theory of Sequential Decision Making in Primal-Dual Spaces. arXiv preprint arXiv:1405.6757. 2014
- Paper (PDF)
Abstract BibTex

In this paper, we set forth a new vision of reinforcement learning developed by us over the past few years, one that yields mathematically rigorous solutions to longstanding important questions that have remained unresolved: (i) how to design reliable, convergent, and robust reinforcement learning algorithms (ii) how to guarantee that reinforcement learning satisfies pre-specified "safety" guarantees, and remains in a stable region of the parameter space (iii) how to design "off-policy" temporal difference learning algorithms in a reliable and stable manner, and finally (iv) how to integrate the study of reinforcement learning into the rich theory of stochastic optimization. In this paper, we provide detailed answers to all these questions using the powerful framework of proximal operators. The key idea that emerges is the use of primal dual spaces connected through the use of a Legendre transform. This allows temporal difference updates to occur in dual spaces, allowing a variety of important technical advantages. The Legendre transform elegantly generalizes past algorithms for solving reinforcement learning problems, such as natural gradient methods, which we show relate closely to the previously unconnected framework of mirror descent methods. Equally importantly, proximal operator theory enables the systematic development of operator splitting methods that show how to safely and reliably decompose complex products of gradients that occur in recent variants of gradient-based temporal difference learning. This key technical innovation makes it possible to finally design "true" stochastic gradient methods for reinforcement learning. Finally, Legendre transforms enable a variety of other benefits, including modeling sparsity and domain geometry. Our work builds extensively on recent work on the convergence of saddle-point algorithms, and on the theory of monotone operators.
```
          @article{mahadevan2014proximal,
            title={Proximal Reinforcement Learning: A New Theory of Sequential Decision Making in Primal-Dual Spaces},
            author={Mahadevan, Sridhar and Liu, Bo and Thomas, Philip and Dabney, Will and Giguere, Steve and Jacek, Nicholas and Gemp, Ian and Liu, Ji},
            journal={arXiv preprint arXiv:1405.6757},
            year={2014}
          }
          
```

2011

I. Gemp, R. Carthew, S. Hilgenfeldt. Cadherin-dependent Cell Morphology in an Epithelium: constructing a quantitative dynamical model. PLoS Computational Biology. 2011
- Paper (PDF)
Abstract BibTex

Cells in the Drosophila retina have well-defined morphologies that are attained during tissue morphogenesis. We present a computer simulation of the epithelial tissue in which the global interfacial energy between cells is minimized. Experimental data for both normal cells and mutant cells either lacking or misexpressing the adhesion protein N-cadherin can be explained by a simple model incorporating salient features of morphogenesis that include the timing of N-cadherin expression in cells and its temporal relationship to the remodeling of cell-cell contacts. The simulations reproduce the geometries of wild-type and mutant cells, distinguish features of cadherin dynamics, and emphasize the importance of adhesion protein biogenesis and its timing with respect to cell remodeling. The simulations also indicate that N-cadherin protein is recycled from inactive interfaces to active interfaces, thereby modulating adhesion strengths between cells.
```
          @article{gemp2011cadherin,
            title={Cadherin-dependent Cell Morphology in an Epithelium: constructing a quantitative dynamical model},
            author={Gemp, Ian M and Carthew, Richard W and Hilgenfeldt, Sascha},
            journal={PLoS computational biology},
            volume={7},
            number={7},
            pages={e1002115},
            publisher={Public Library of Science},
            year={2011}
          }
          
```

2010

G. Duncan, K. Rabl, I. Gemp, R. Heidelberger, W.B. Thoreson. Quantitative Analysis of Synaptic Release at the Photoreceptor Synapse. Biophysical Journal. 2010
- Paper (PDF)
Abstract BibTex

Exocytosis from the rod photoreceptor is stimulated by submicromolar Ca2+ and exhibits an unusually shallow dependence on presynaptic Ca2+. To provide a quantitative description of the photoreceptor Ca2+ sensor for exocytosis, we tested a family of conventional and allosteric computational models describing the final Ca2+-binding steps leading to exocytosis. Simulations were fit to two measures of release, evoked by flash-photolysis of caged Ca2+: exocytotic capacitance changes from individual rods and postsynaptic currents of second-order neurons. The best simulations supported the occupancy of only two Ca2+ binding sites on the rod Ca2+ sensor rather than the typical four or five. For most models, the on-rates for Ca2+ binding and maximal fusion rate were comparable to those of other neurons. However, the off-rates for Ca2+ unbinding were unexpectedly slow. In addition to contributing to the high-affinity of the photoreceptor Ca2+ sensor, slow Ca2+ unbinding may support the fusion of vesicles located at a distance from Ca2+ channels. In addition, partial sensor occupancy due to slow unbinding may contribute to the linearization of the first synapse in vision.
```
          @article{duncan2010quantitative,
            title={Quantitative Analysis of Synaptic Release at the Photoreceptor Synapse},
            author={Duncan, Gabriel and Rabl, Katalin and Gemp, Ian and Heidelberger, Ruth and Thoreson, Wallace B},
            journal={Biophysical journal},
            volume={98},
            number={10},
            pages={2102--2110},
            publisher={Elsevier},
            year={2010}
          }
          
```