-
Generative AI Models: Opportunities and Risks for Industry and Authorities
Authors:
Tobias Alt,
Andrea Ibisch,
Clemens Meiser,
Anna Wilhelm,
Raphael Zimmer,
Christian Berghoff,
Christoph Droste,
Jens Karschau,
Friederike Laus,
Rainer Plaga,
Carola Plesch,
Britta Sennewald,
Thomas Thaeren,
Kristina Unverricht,
Steffen Waurick
Abstract:
Generative AI models are capable of performing a wide range of tasks that traditionally require creativity and human understanding. They learn patterns from existing data during training and can subsequently generate new content such as texts, images, and music that follow these patterns. Due to their versatility and generally high-quality results, they, on the one hand, represent an opportunity f…
▽ More
Generative AI models are capable of performing a wide range of tasks that traditionally require creativity and human understanding. They learn patterns from existing data during training and can subsequently generate new content such as texts, images, and music that follow these patterns. Due to their versatility and generally high-quality results, they, on the one hand, represent an opportunity for digitalization. On the other hand, the use of generative AI models introduces novel IT security risks that need to be considered for a comprehensive analysis of the threat landscape in relation to IT security. In response to this risk potential, companies or authorities using them should conduct an individual risk analysis before integrating generative AI into their workflows. The same applies to developers and operators, as many risks in the context of generative AI have to be taken into account at the time of development or can only be influenced by the operating company. Based on this, existing security measures can be adjusted, and additional measures can be taken.
△ Less
Submitted 7 June, 2024;
originally announced June 2024.
-
Hybrid Models for Facial Emotion Recognition in Children
Authors:
Rafael Zimmer,
Marcos Sobral,
Helio Azevedo
Abstract:
This paper focuses on the use of emotion recognition techniques to assist psychologists in performing children's therapy through remotely robot operated sessions. In the field of psychology, the use of agent-mediated therapy is growing increasingly given recent advances in robotics and computer science. Specifically, the use of Embodied Conversational Agents (ECA) as an intermediary tool can help…
▽ More
This paper focuses on the use of emotion recognition techniques to assist psychologists in performing children's therapy through remotely robot operated sessions. In the field of psychology, the use of agent-mediated therapy is growing increasingly given recent advances in robotics and computer science. Specifically, the use of Embodied Conversational Agents (ECA) as an intermediary tool can help professionals connect with children who face social challenges such as Attention Deficit Hyperactivity Disorder (ADHD), Autism Spectrum Disorder (ASD) or even who are physically unavailable due to being in regions of armed conflict, natural disasters, or other circumstances. In this context, emotion recognition represents an important feedback for the psychotherapist. In this article, we initially present the result of a bibliographical research associated with emotion recognition in children. This research revealed an initial overview on algorithms and datasets widely used by the community. Then, based on the analysis carried out on the results of the bibliographical research, we used the technique of dense optical flow features to improve the ability of identifying emotions in children in uncontrolled environments. From the output of a hybrid model of Convolutional Neural Network, two intermediary features are fused before being processed by a final classifier. The proposed architecture was called HybridCNNFusion. Finally, we present the initial results achieved in the recognition of children's emotions using a dataset of Brazilian children.
△ Less
Submitted 24 August, 2023;
originally announced August 2023.
-
Encrypted Internet traffic classification using a supervised Spiking Neural Network
Authors:
Ali Rasteh,
Florian Delpech,
Carlos Aguilar-Melchor,
Romain Zimmer,
Saeed Bagheri Shouraki,
Timothée Masquelier
Abstract:
Internet traffic recognition is an essential tool for access providers since recognizing traffic categories related to different data packets transmitted on a network help them define adapted priorities. That means, for instance, high priority requirements for an audio conference and low ones for a file transfer, to enhance user experience. As internet traffic becomes increasingly encrypted, the m…
▽ More
Internet traffic recognition is an essential tool for access providers since recognizing traffic categories related to different data packets transmitted on a network help them define adapted priorities. That means, for instance, high priority requirements for an audio conference and low ones for a file transfer, to enhance user experience. As internet traffic becomes increasingly encrypted, the mainstream classic traffic recognition technique, payload inspection, is rendered ineffective. This paper uses machine learning techniques for encrypted traffic classification, looking only at packet size and time of arrival. Spiking neural networks (SNN), largely inspired by how biological neurons operate, were used for two reasons. Firstly, they are able to recognize time-related data packet features. Secondly, they can be implemented efficiently on neuromorphic hardware with a low energy footprint. Here we used a very simple feedforward SNN, with only one fully-connected hidden layer, and trained in a supervised manner using the newly introduced method known as Surrogate Gradient Learning. Surprisingly, such a simple SNN reached an accuracy of 95.9% on ISCX datasets, outperforming previous approaches. Besides better accuracy, there is also a very significant improvement on simplicity: input size, number of neurons, trainable parameters are all reduced by one to four orders of magnitude. Next, we analyzed the reasons for this good accuracy. It turns out that, beyond spatial (i.e. packet size) features, the SNN also exploits temporal ones, mostly the nearly synchronous (within a 200ms range) arrival times of packets with certain sizes. Taken together, these results show that SNNs are an excellent fit for encrypted internet traffic classification: they can be more accurate than conventional artificial neural networks (ANN), and they could be implemented efficiently on low power embedded systems.
△ Less
Submitted 21 July, 2022; v1 submitted 24 January, 2021;
originally announced January 2021.
-
Low-activity supervised convolutional spiking neural networks applied to speech commands recognition
Authors:
Thomas Pellegrini,
Romain Zimmer,
Timothée Masquelier
Abstract:
Deep Neural Networks (DNNs) are the current state-of-the-art models in many speech related tasks. There is a growing interest, though, for more biologically realistic, hardware friendly and energy efficient models, named Spiking Neural Networks (SNNs). Recently, it has been shown that SNNs can be trained efficiently, in a supervised manner, using backpropagation with a surrogate gradient trick. In…
▽ More
Deep Neural Networks (DNNs) are the current state-of-the-art models in many speech related tasks. There is a growing interest, though, for more biologically realistic, hardware friendly and energy efficient models, named Spiking Neural Networks (SNNs). Recently, it has been shown that SNNs can be trained efficiently, in a supervised manner, using backpropagation with a surrogate gradient trick. In this work, we report speech command (SC) recognition experiments using supervised SNNs. We explored the Leaky-Integrate-Fire (LIF) neuron model for this task, and show that a model comprised of stacked dilated convolution spiking layers can reach an error rate very close to standard DNNs on the Google SC v1 dataset: 5.5%, while kee** a very sparse spiking activity, below 5%, thank to a new regularization term. We also show that modeling the leakage of the neuron membrane potential is useful, since the LIF model outperformed its non-leaky model counterpart significantly.
△ Less
Submitted 13 November, 2020;
originally announced November 2020.
-
Wavefront sensing and control in space-based coronagraph instruments using Zernike's phase-contrast method
Authors:
Garreth Ruane,
J. Kent Wallace,
John Steeves,
Camilo Mejia Prada,
Byoung-Joon Seo,
Eduardo Bendek,
Carl Coker,
Pin Chen,
Brendan Crill,
Jeff Jewell,
Brian Kern,
David Marx,
Phillip K. Poon,
David Redding,
A J Eldorado Riggs,
Nicholas Siegler,
Robert Zimmer
Abstract:
Future space telescopes with coronagraph instruments will use a wavefront sensor (WFS) to measure and correct for phase errors and stabilize the stellar intensity in high-contrast images. The HabEx and LUVOIR mission concepts baseline a Zernike wavefront sensor (ZWFS), which uses Zernike's phase contrast method to convert phase in the pupil into intensity at the WFS detector. In preparation for th…
▽ More
Future space telescopes with coronagraph instruments will use a wavefront sensor (WFS) to measure and correct for phase errors and stabilize the stellar intensity in high-contrast images. The HabEx and LUVOIR mission concepts baseline a Zernike wavefront sensor (ZWFS), which uses Zernike's phase contrast method to convert phase in the pupil into intensity at the WFS detector. In preparation for these potential future missions, we experimentally demonstrate a ZWFS in a coronagraph instrument on the Decadal Survey Testbed in the High Contrast Imaging Testbed facility at NASA's Jet Propulsion Laboratory. We validate that the ZWFS can measure low- and mid-spatial frequency aberrations up to the control limit of the deformable mirror, with surface height sensitivity as small as 1 pm, using a configuration similar to the HabEx and LUVOIR concepts. Furthermore, we demonstrate closed-loop control, resolving an individual DM actuator, with residuals consistent with theoretical models. In addition, we predict the expected performance of a ZWFS on future space telescopes using natural starlight from a variety of spectral types. The most challenging scenarios require ~1 hr of integration time to achieve picometer sensitivity. This timescale may be drastically reduced by using internal or external laser sources for sensing purposes. The experimental results and theoretical predictions presented here advance the WFS technology in the context of the next generation of space telescopes with coronagraph instruments.
△ Less
Submitted 3 November, 2020; v1 submitted 20 October, 2020;
originally announced October 2020.
-
Technical report: supervised training of convolutional spiking neural networks with PyTorch
Authors:
Romain Zimmer,
Thomas Pellegrini,
Srisht Fateh Singh,
Timothée Masquelier
Abstract:
Recently, it has been shown that spiking neural networks (SNNs) can be trained efficiently, in a supervised manner, using backpropagation through time. Indeed, the most commonly used spiking neuron model, the leaky integrate-and-fire neuron, obeys a differential equation which can be approximated using discrete time steps, leading to a recurrent relation for the potential. The firing threshold cau…
▽ More
Recently, it has been shown that spiking neural networks (SNNs) can be trained efficiently, in a supervised manner, using backpropagation through time. Indeed, the most commonly used spiking neuron model, the leaky integrate-and-fire neuron, obeys a differential equation which can be approximated using discrete time steps, leading to a recurrent relation for the potential. The firing threshold causes optimization issues, but they can be overcome using a surrogate gradient. Here, we extend previous approaches in two ways. Firstly, we show that the approach can be used to train convolutional layers. Convolutions can be done in space, time (which simulates conduction delays), or both. Secondly, we include fast horizontal connections à la Denève: when a neuron N fires, we subtract to the potentials of all the neurons with the same receptive the dot product between their weight vectors and the one of neuron N. As Denève et al. showed, this is useful to represent a dynamic multidimensional analog signal in a population of spiking neurons. Here we demonstrate that, in addition, such connections also allow implementing a multidimensional send-on-delta coding scheme. We validate our approach on one speech classification benchmarks: the Google speech command dataset. We managed to reach nearly state-of-the-art accuracy (94%) while maintaining low firing rates (about 5Hz). Our code is based on PyTorch and is available in open source at http://github.com/romainzimmer/s2net
△ Less
Submitted 22 November, 2019;
originally announced November 2019.
-
An Elementary Approach To Uniform In Time Propagation Of Chaos
Authors:
Alain Durmus,
Andreas Eberle,
Arnaud Guillin,
Raphael Zimmer
Abstract:
Based on a coupling approach, we prove uniform in time propagation of chaos for weakly interacting mean-field particle systems with possibly non-convex confinement and interaction potentials. The approach is based on a combination of reflection and synchronous couplings applied to the individual particles. It provides explicit quantitative bounds that significantly extend previous results for the…
▽ More
Based on a coupling approach, we prove uniform in time propagation of chaos for weakly interacting mean-field particle systems with possibly non-convex confinement and interaction potentials. The approach is based on a combination of reflection and synchronous couplings applied to the individual particles. It provides explicit quantitative bounds that significantly extend previous results for the convex case.
△ Less
Submitted 29 May, 2018;
originally announced May 2018.
-
Coupling and Convergence for Hamiltonian Monte Carlo
Authors:
Nawaf Bou-Rabee,
Andreas Eberle,
Raphael Zimmer
Abstract:
Based on a new coupling approach, we prove that the transition step of the Hamiltonian Monte Carlo algorithm is contractive w.r.t. a carefully designed Kantorovich (L1 Wasserstein) distance. The lower bound for the contraction rate is explicit. Global convexity of the potential is not required, and thus multimodal target distributions are included. Explicit quantitative bounds for the number of st…
▽ More
Based on a new coupling approach, we prove that the transition step of the Hamiltonian Monte Carlo algorithm is contractive w.r.t. a carefully designed Kantorovich (L1 Wasserstein) distance. The lower bound for the contraction rate is explicit. Global convexity of the potential is not required, and thus multimodal target distributions are included. Explicit quantitative bounds for the number of steps required to approximate the stationary distribution up to a given error are a direct consequence of contractivity. These bounds show that HMC can overcome diffusive behaviour if the duration of the Hamiltonian dynamics is adjusted appropriately.
△ Less
Submitted 29 May, 2019; v1 submitted 1 May, 2018;
originally announced May 2018.
-
Couplings and quantitative contraction rates for Langevin dynamics
Authors:
Andreas Eberle,
Arnaud Guillin,
Raphael Zimmer
Abstract:
We introduce a new probabilistic approach to quantify convergence to equilibrium for (kinetic) Langevin processes. In contrast to previous analytic approaches that focus on the associated kinetic Fokker-Planck equation, our approach is based on a specific combination of reflection and synchronous coupling of two solutions of the Langevin equation. It yields contractions in a particular Wasserstein…
▽ More
We introduce a new probabilistic approach to quantify convergence to equilibrium for (kinetic) Langevin processes. In contrast to previous analytic approaches that focus on the associated kinetic Fokker-Planck equation, our approach is based on a specific combination of reflection and synchronous coupling of two solutions of the Langevin equation. It yields contractions in a particular Wasserstein distance, and it provides rather precise bounds for convergence to equilibrium at the borderline between the overdamped and the underdamped regime. In particular, we are able to recover kinetic behavior in terms of explicit lower bounds for the contraction rate. For example, for a rescaled double-well potential with local minima at distance $a$, we obtain a lower bound for the contraction rate of order $Ω(a^{-1})$ provided the friction coefficient is of order $Θ(a^{-1})$.
△ Less
Submitted 29 June, 2018; v1 submitted 5 March, 2017;
originally announced March 2017.
-
Sticky couplings of multidimensional diffusions with different drifts
Authors:
Andreas Eberle,
Raphael Zimmer
Abstract:
We present a novel approach of coupling two multidimensional and non-degenerate Itô processes $(X_t)$ and $(Y_t)$ which follow dynamics with different drifts. Our coupling is sticky in the sense that there is a stochastic process $(r_t)$, which solves a one-dimensional stochastic differential equation with a sticky boundary behavior at zero, such that almost surely $|X_t-Y_t|\leq r_t$ for all…
▽ More
We present a novel approach of coupling two multidimensional and non-degenerate Itô processes $(X_t)$ and $(Y_t)$ which follow dynamics with different drifts. Our coupling is sticky in the sense that there is a stochastic process $(r_t)$, which solves a one-dimensional stochastic differential equation with a sticky boundary behavior at zero, such that almost surely $|X_t-Y_t|\leq r_t$ for all $t\geq 0$. The coupling is constructed as a weak limit of Markovian couplings. We provide explicit, non-asymptotic and long-time stable bounds for the probability of the event $\{X_t=Y_t\}$.
△ Less
Submitted 19 December, 2016;
originally announced December 2016.
-
Quantitative Harris type theorems for diffusions and McKean-Vlasov processes
Authors:
Andreas Eberle,
Arnaud Guillin,
Raphael Zimmer
Abstract:
We consider $\mathbb{R}^d$-valued diffusion processes of type
\begin{align*}
dX_t\ =\ b(X_t)dt\, +\, dB_t.
\end{align*} Assuming a geometric drift condition, we establish contractions of the transitions kernels in Kantorovich ($L^1$ Wasserstein) distances with explicit constants. Our results are in the spirit of Hairer and Mattingly's extension of Harris' Theorem. In particular, they do not…
▽ More
We consider $\mathbb{R}^d$-valued diffusion processes of type
\begin{align*}
dX_t\ =\ b(X_t)dt\, +\, dB_t.
\end{align*} Assuming a geometric drift condition, we establish contractions of the transitions kernels in Kantorovich ($L^1$ Wasserstein) distances with explicit constants. Our results are in the spirit of Hairer and Mattingly's extension of Harris' Theorem. In particular, they do not rely on a small set condition. Instead we combine Lyapunov functions with reflection coupling and concave distance functions. We retrieve constants that are explicit in parameters which can be computed with little effort from one-sided Lipschitz conditions for the drift coefficient and the growth of a chosen Lyapunov function. Consequences include exponential convergence in weighted total variation norms, gradient bounds, bounds for ergodic averages, and Kantorovich contractions for nonlinear McKean-Vlasov diffusions in the case of sufficiently weak but not necessarily bounded nonlinearities. We also establish quantitative bounds for sub-geometric ergodicity assuming a sub-geometric drift condition.
△ Less
Submitted 9 October, 2017; v1 submitted 20 June, 2016;
originally announced June 2016.
-
Explicit contraction rates for a class of degenerate and infinite-dimensional diffusions
Authors:
Raphael Zimmer
Abstract:
Given a separable and real Hilbert space $\mathbb{H}$ and a trace-class, symmetric and non-negative operator $\mathcal{G}:\mathbb{H}\rightarrow\mathbb{H}$, we examine the equation \begin{align*}
dX_t = -X_t\, dt + b(X_t) \, dt + \sqrt{2} \, dW_t, \qquad X_0=x\in\mathbb{H},
\end{align*} where $(W_t)$ is a $\mathcal{G}$-Wiener process on $\mathbb{H}$ and $b:\mathbb{H}\rightarrow\mathbb{H}$ is Li…
▽ More
Given a separable and real Hilbert space $\mathbb{H}$ and a trace-class, symmetric and non-negative operator $\mathcal{G}:\mathbb{H}\rightarrow\mathbb{H}$, we examine the equation \begin{align*}
dX_t = -X_t\, dt + b(X_t) \, dt + \sqrt{2} \, dW_t, \qquad X_0=x\in\mathbb{H},
\end{align*} where $(W_t)$ is a $\mathcal{G}$-Wiener process on $\mathbb{H}$ and $b:\mathbb{H}\rightarrow\mathbb{H}$ is Lipschitz. We assume there is a splitting of $\mathbb{H}$ into a finite-dimensional space $\mathbb{H}^l$ and its orthogonal complement $\mathbb{H}^h$ such that $\mathcal{G}$ is strictly positive definite on $\mathbb{H}^l$ and the non-linearity $b$ admits a contraction property on $\mathbb{H}^h$. Assuming a geometric drift condition, we derive a Kantorovich ($L^1$ Wasserstein) contraction with an explicit rate for the corresponding Markov kernels. The estimates for the rate are based on the eigenvalues of $\mathcal{G}$ on the space $\mathbb{H}^l$, a Lipschitz bound on $b$ and a geometric drift condition. The results are derived using coupling methods.
△ Less
Submitted 31 January, 2017; v1 submitted 25 May, 2016;
originally announced May 2016.
-
Demonstration of high contrast with an obscured aperture with the WFIRST-AFTA shaped pupil coronagraph
Authors:
Eric Cady,
Camilo Mejia Prada,
Xin An,
Kunjithapatham Balasubramanian,
Rosemary Diaz,
N. Jeremy Kasdin,
Brian Kern,
Andreas Kuhnert,
Bijan Nemati,
Ilya Poberezhskiy,
A. J. Eldorado Riggs,
Robert Zimmer,
Neil Zimmerman
Abstract:
The coronagraph instrument on the WFIRST-AFTA mission study has two coronagraphic architectures, shaped pupil and hybrid Lyot, which may be interchanged for use in different observing scenarios. Each architecture relies on newly-developed mask components to function in the presence of the AFTA aperture, and so both must be matured to a high technology readiness level (TRL) in advance of the missio…
▽ More
The coronagraph instrument on the WFIRST-AFTA mission study has two coronagraphic architectures, shaped pupil and hybrid Lyot, which may be interchanged for use in different observing scenarios. Each architecture relies on newly-developed mask components to function in the presence of the AFTA aperture, and so both must be matured to a high technology readiness level (TRL) in advance of the mission. A series of milestones were set to track the development of the technologies required for the instrument; in this paper, we report on completion of WFIRST-AFTA Coronagraph Milestone 2---a narrowband $10^{-8}$ contrast test with static aberrations for the shaped pupil---and the plans for the upcoming broadband Coronagraph Milestone 5.
△ Less
Submitted 5 November, 2015;
originally announced November 2015.
-
Volume-preserving actions of simple algebraic Q-groups on low-dimensional manifolds
Authors:
Dave Witte Morris,
Robert J. Zimmer
Abstract:
We prove that SL(n,Q) has no nontrivial, C-infinity, volume-preserving action on any compact manifold of dimension strictly less than n. More generally, suppose G is a connected, isotropic, almost-simple algebraic group over Q, such that the simple factors of every localization of G have rank at least two. If there does not exist a nontrivial homomorphism from G(R) to GL(d,C), then every C-infinit…
▽ More
We prove that SL(n,Q) has no nontrivial, C-infinity, volume-preserving action on any compact manifold of dimension strictly less than n. More generally, suppose G is a connected, isotropic, almost-simple algebraic group over Q, such that the simple factors of every localization of G have rank at least two. If there does not exist a nontrivial homomorphism from G(R) to GL(d,C), then every C-infinity, volume-preserving action of G(Q) on any compact d-dimensional manifold must factor through a finite group.
△ Less
Submitted 14 April, 2012;
originally announced April 2012.
-
Chemotactic Maneuverability of Sperm
Authors:
Jeffrey S. Guasto,
Jeffrey A. Riffell,
Richard K. Zimmer,
Roman Stocker
Abstract:
In this fluid mechanics video, we explore the kinematics of chemotaxing sperm cells (sea urchin, \textit{Arbacia punctulata}) swimming in a chemoattractant gradient. We demonstrate that the complex swimming trajectories resulting in chemotactic behavior (`turn-and-run' motility) are comprised of several distinct flagellar maneuvers. These motility patterns likely play an important role optimizing…
▽ More
In this fluid mechanics video, we explore the kinematics of chemotaxing sperm cells (sea urchin, \textit{Arbacia punctulata}) swimming in a chemoattractant gradient. We demonstrate that the complex swimming trajectories resulting in chemotactic behavior (`turn-and-run' motility) are comprised of several distinct flagellar maneuvers. These motility patterns likely play an important role optimizing chemotaxic motility and navigation, when the sperm cells are subjected external fluid flows.
△ Less
Submitted 13 October, 2011;
originally announced October 2011.
-
Invariant rigid geometric structures and smooth projective factors
Authors:
Amos Nevo,
Robert J. Zimmer
Abstract:
We consider actions of non-compact simple Lie groups preserving an analytic rigid geometric structure of algebraic type on a compact manifold. The structure is not assumed to be unimodular, so an invariant measure may not exist. Ergodic stationary measures always exist, and when such a measure has full support, we show the following.
1) Either the manifold admits a smooth equivariant map onto…
▽ More
We consider actions of non-compact simple Lie groups preserving an analytic rigid geometric structure of algebraic type on a compact manifold. The structure is not assumed to be unimodular, so an invariant measure may not exist. Ergodic stationary measures always exist, and when such a measure has full support, we show the following.
1) Either the manifold admits a smooth equivariant map onto a homogeneous projective variety, defined on an open dense conull invariant set, or the Lie algebra of the Zariski closure of the Gromov representation of the fundamental group contains a Lie subalgebra isomorphic to the Lie algebra of the acting group. As a corollary, a smooth non-trivial homogeneous projective factor does exist whenever the fundamental group of $M$ admits only virtually solvable linear representations, and thus in particular when $M$ is simply connected, regardless of the real rank.
2) There exist explicit examples showing that analytic rigid actions of certain simple groups (of real rank one) may indeed fail to have a smooth projective factor.
3) It is possible to generalize Gromov's theorem on the algebraic hull of the representation of the fundamental group of the manifold to the case of analytic rigid non-unimodular structures, for actions of simple groups of any real rank.
An important ingredient in the proofs is a generalization of Gromov's centralizer theorem beyond the case of invariant measures.
△ Less
Submitted 6 January, 2009; v1 submitted 6 August, 2007;
originally announced August 2007.
-
Probabilistic methods for predicting protein functions in protein-protein interaction networks
Authors:
Christoph Best,
Ralf Zimmer,
Joannis Apostolakis
Abstract:
We discuss probabilistic methods for predicting protein functions from protein-protein interaction networks. Previous work based on Markov Randon Fields is extended and compared to a general machine-learning theoretic approach. Using actual protein interaction networks for yeast from the MIPS database and GO-SLIM function assignments, we compare the predictions of the different probabilistic met…
▽ More
We discuss probabilistic methods for predicting protein functions from protein-protein interaction networks. Previous work based on Markov Randon Fields is extended and compared to a general machine-learning theoretic approach. Using actual protein interaction networks for yeast from the MIPS database and GO-SLIM function assignments, we compare the predictions of the different probabilistic methods and of a standard support vector machine. It turns out that, with the currently available networks, the simple methods based on counting frequencies perform as well as the more sophisticated approaches.
△ Less
Submitted 12 March, 2005;
originally announced March 2005.
-
Ergodic actions of semisimple Lie groups on compact principal bundles
Authors:
Dave Witte,
Robert J. Zimmer
Abstract:
Let G = SL(n,R) (or, more generally, let G be a connected, noncompact, simple Lie group). For any compact Lie group K, it is easy to find a compact manifold M, such that there is a volume-preserving, connection-preserving, ergodic action of G on some smooth, principal K-bundle P over M. Can M can be chosen independent of K? We show that if M = H/L is a homogeneous space, and the action of G on M…
▽ More
Let G = SL(n,R) (or, more generally, let G be a connected, noncompact, simple Lie group). For any compact Lie group K, it is easy to find a compact manifold M, such that there is a volume-preserving, connection-preserving, ergodic action of G on some smooth, principal K-bundle P over M. Can M can be chosen independent of K? We show that if M = H/L is a homogeneous space, and the action of G on M is by translations, then P must also be a homogeneous space H'/L'. Consequently, there is a strong restriction on the groups K that can arise over this particular M.
△ Less
Submitted 30 May, 2002;
originally announced May 2002.
-
Actions of semisimple Lie groups on circle bundles
Authors:
Dave Witte,
Robert J. Zimmer
Abstract:
Suppose G is a connected, simple, real Lie group with real rank at least two, M is an ergodic G-space with invariant probability measure, and f is a Homeo(T)-valued Borel cocycle, where Homeo(T) denotes the group of homeomorphisms of the circle T. We use an argument of E.Ghys to show that there is a G-invariant probability measure on the skew product of M and T. Furthermore, if the image of f co…
▽ More
Suppose G is a connected, simple, real Lie group with real rank at least two, M is an ergodic G-space with invariant probability measure, and f is a Homeo(T)-valued Borel cocycle, where Homeo(T) denotes the group of homeomorphisms of the circle T. We use an argument of E.Ghys to show that there is a G-invariant probability measure on the skew product of M and T. Furthermore, if the image of f consists of diffeomorphisms, then there is an invariant measure that is equivalent to the product measure; therefore, f is cohomologous to a cocycle with values in the isometry group of T.
△ Less
Submitted 10 April, 2000;
originally announced April 2000.
