-
Heegaard Floer invariants for cyclic 3-orbifolds
Authors:
Saibal Ganguli,
Mainak Poddar
Abstract:
We define a notion of Heegaard Floer homology for three dimensional orbifolds with arbitrary cyclic singularities, generalizing the recent work of Biji Wong where the singular locus is assumed to be connected.
We define a notion of Heegaard Floer homology for three dimensional orbifolds with arbitrary cyclic singularities, generalizing the recent work of Biji Wong where the singular locus is assumed to be connected.
△ Less
Submitted 13 February, 2024; v1 submitted 22 October, 2022;
originally announced October 2022.
-
Linear Stability Analysis of a Normal Shock Train in a Constant Area Isolator of a Hypersonic ScramJet
Authors:
Swetava Ganguli
Abstract:
Recent experiments in hypersonic vehicle technology have had some successful outcomes (ex. NASA X-43, Boeing X-51) but also have seen some unexpected failures (DARPA Hypersonic Test Vehicle 2). One important failure mode is the problem of engine unstart. Morgan, et al (2012) has discussed and analyzed a Scramjet isolator using two unit problems, viz. that of an oblique shock im**ing on a turbule…
▽ More
Recent experiments in hypersonic vehicle technology have had some successful outcomes (ex. NASA X-43, Boeing X-51) but also have seen some unexpected failures (DARPA Hypersonic Test Vehicle 2). One important failure mode is the problem of engine unstart. Morgan, et al (2012) has discussed and analyzed a Scramjet isolator using two unit problems, viz. that of an oblique shock im**ing on a turbulent boundary layer and that of a normal shock train in a constant-area duct. In order to analyze unstart, we use the numerical data obtained from the LES calculations of the normal shock train in a constant area duct and perform temporal linear stability analysis on it under the assumptions of inviscid semi-parallel flow. This paper describes the methodology and the details of this linear stability analysis. The results obtained from the investigations have been documented and interpreted with their relevance to the physical problem. The report concludes that temporal linear stability analysis is insufficient to analyze the problem of unstart. Finally, the need for a new approach (namely, dynamic mode decomposition (DMD)) has been discussed.
△ Less
Submitted 30 July, 2019; v1 submitted 19 July, 2019;
originally announced July 2019.
-
Determinat Bundles and Geometric Quantization Of Vortex Moduli Spaces ON Compact Kahler Surfaces
Authors:
Saibal Ganguli
Abstract:
In this paper we first show that on projective manifolds (M, ω), there are holomorphic determinant bundles (in the sense of Knusden-Mumford used by Bismut, Gillet, Soule) which play the role of the geometric quantum bundle, namely one for each input data of a Hermitian holomorphic line bundle L of non-trivial Chern class on a compact Kahler manifold Z (with Todd genus non-zero) and a choice of a g…
▽ More
In this paper we first show that on projective manifolds (M, ω), there are holomorphic determinant bundles (in the sense of Knusden-Mumford used by Bismut, Gillet, Soule) which play the role of the geometric quantum bundle, namely one for each input data of a Hermitian holomorphic line bundle L of non-trivial Chern class on a compact Kahler manifold Z (with Todd genus non-zero) and a choice of a geometric quantization of (M, ω). Next we further study the generalization of the vortex equations on Kahler 4-manifold which has been studied earlier by Bradlow. We show that when the Kahler 4-manifold avoids some obstructions then the regular part of the moduli space is a Kahler manifold and admit a pull back of a Quillen determinant bundle as the quantum line bundle, i.e. the curvature is proportional to the Kahler form. Thus they can be quantized geometrically. In fact we show that the moduli space of the usual vortex equations on a projective Kahler 4-manifold is projective when the moduli space is smooth. Since in Kahler 4-manifold the vortex moduli and the Seiberg Witten moduli coincide our effort gives a quantization of Seiberg Witten moduli by determinant bundles
△ Less
Submitted 14 June, 2021; v1 submitted 8 August, 2018;
originally announced August 2018.
-
Geometric quantization of finite Toda systems and coherent States
Authors:
Rukmini Dey,
Saibal Ganguli
Abstract:
Adler had shown in 1979 that the Toda system can be given a coad- joint orbit description. We quantize the Toda system by viewing it as a single orbit of a multiplicative group of lower triangular matrices of determinant one with pos- itive diagonal entries. We get a unitary representation of the group with square integrable polarized sections of the quantization as the module . We find the Rawnsl…
▽ More
Adler had shown in 1979 that the Toda system can be given a coad- joint orbit description. We quantize the Toda system by viewing it as a single orbit of a multiplicative group of lower triangular matrices of determinant one with pos- itive diagonal entries. We get a unitary representation of the group with square integrable polarized sections of the quantization as the module . We find the Rawnsley coherent states after a completion of the above space of sections. We also find non-unitary finite dimensional quantum Hilbert spaces for the system.
△ Less
Submitted 17 April, 2017; v1 submitted 9 December, 2016;
originally announced December 2016.
-
An equivalence between high dimensional Bayes optimal inference and M-estimation
Authors:
Madhu Advani,
Surya Ganguli
Abstract:
When recovering an unknown signal from noisy measurements, the computational difficulty of performing optimal Bayesian MMSE (minimum mean squared error) inference often necessitates the use of maximum a posteriori (MAP) inference, a special case of regularized M-estimation, as a surrogate. However, MAP is suboptimal in high dimensions, when the number of unknown signal components is similar to the…
▽ More
When recovering an unknown signal from noisy measurements, the computational difficulty of performing optimal Bayesian MMSE (minimum mean squared error) inference often necessitates the use of maximum a posteriori (MAP) inference, a special case of regularized M-estimation, as a surrogate. However, MAP is suboptimal in high dimensions, when the number of unknown signal components is similar to the number of measurements. In this work we demonstrate, when the signal distribution and the likelihood function associated with the noise are both log-concave, that optimal MMSE performance is asymptotically achievable via another M-estimation procedure. This procedure involves minimizing convex loss and regularizer functions that are nonlinearly smoothed versions of the widely applied MAP optimization problem. Our findings provide a new heuristic derivation and interpretation for recent optimal M-estimators found in the setting of linear measurements and additive noise, and further extend these results to nonlinear measurements with non-additive noise. We numerically demonstrate superior performance of our optimal M-estimators relative to MAP. Overall, at the heart of our work is the revelation of a remarkable equivalence between two seemingly very different computational problems: namely that of high dimensional Bayesian integration underlying MMSE inference, and high dimensional convex optimization underlying M-estimation. In essence we show that the former difficult integral may be computed by solving the latter, simpler optimization problem.
△ Less
Submitted 22 September, 2016;
originally announced September 2016.
-
The dimension of the Hilbert Space of Geometic quantization of vortices on a Riemann surface
Authors:
Rukmini Dey,
Saibal Ganguli
Abstract:
In this article we calculate the dimension of the Hilbert space of Kahler quantization of the moduli space of vortices on a Riemann surface. This dimension is given by the holomorphic Euler characteristic of the quantum line bundle.
In this article we calculate the dimension of the Hilbert space of Kahler quantization of the moduli space of vortices on a Riemann surface. This dimension is given by the holomorphic Euler characteristic of the quantum line bundle.
△ Less
Submitted 24 February, 2017; v1 submitted 13 June, 2016;
originally announced June 2016.
-
Statistical Mechanics of High-Dimensional Inference
Authors:
Madhu Advani,
Surya Ganguli
Abstract:
To model modern large-scale datasets, we need efficient algorithms to infer a set of $P$ unknown model parameters from $N$ noisy measurements. What are fundamental limits on the accuracy of parameter inference, given finite signal-to-noise ratios, limited measurements, prior information, and computational tractability requirements? How can we combine prior information with measurements to achieve…
▽ More
To model modern large-scale datasets, we need efficient algorithms to infer a set of $P$ unknown model parameters from $N$ noisy measurements. What are fundamental limits on the accuracy of parameter inference, given finite signal-to-noise ratios, limited measurements, prior information, and computational tractability requirements? How can we combine prior information with measurements to achieve these limits? Classical statistics gives incisive answers to these questions as the measurement density $α= \frac{N}{P}\rightarrow \infty$. However, these classical results are not relevant to modern high-dimensional inference problems, which instead occur at finite $α$. We formulate and analyze high-dimensional inference as a problem in the statistical physics of quenched disorder. Our analysis uncovers fundamental limits on the accuracy of inference in high dimensions, and reveals that widely cherished inference algorithms like maximum likelihood (ML) and maximum-a posteriori (MAP) inference cannot achieve these limits. We further find optimal, computationally tractable algorithms that can achieve these limits. Intriguingly, in high dimensions, these optimal algorithms become computationally simpler than MAP and ML, while still outperforming them. For example, such optimal algorithms can lead to as much as a 20% reduction in the amount of data to achieve the same performance relative to MAP. Moreover, our analysis reveals simple relations between optimal high dimensional inference and low dimensional scalar Bayesian inference, insights into the nature of generalization and predictive power in high dimensions, information theoretic limits on compressed sensing, phase transitions in quadratic inference, and connections to central mathematical objects in convex optimization theory and random matrix theory.
△ Less
Submitted 21 February, 2016; v1 submitted 18 January, 2016;
originally announced January 2016.
-
A classification result and contact structures in oriented cyclic orbifold
Authors:
Saibal Ganguli
Abstract:
We prove every oriented compact cyclic $3$-orbifold has a contact structure. There is another proof in the web by Daniel Herr in his uploaded thesis which depends on open book decompositions, ours is independent of that. We define overtwisted contact structures, tight contact structures and Lutz twist on oriented compact cyclic 3-orbifolds. We show every contact structure in an oriented compact cy…
▽ More
We prove every oriented compact cyclic $3$-orbifold has a contact structure. There is another proof in the web by Daniel Herr in his uploaded thesis which depends on open book decompositions, ours is independent of that. We define overtwisted contact structures, tight contact structures and Lutz twist on oriented compact cyclic 3-orbifolds. We show every contact structure in an oriented compact cyclic $3$-orbifold contactified by our method is homotopic to an overtwisted structure with the overtwisted disc intersecting the singular locus of the orbifold. We pose Eliashberg's like characterization of overtwisted contact structures of cyclic $3$-orbifolds as an open problem. In course of proving the above results we prove a classification result for compact oriented cyclic-3 orbifolds which has not been seen by us in literature before.
△ Less
Submitted 23 December, 2015; v1 submitted 19 March, 2015;
originally announced March 2015.
-
Hodge structures Orbifold hodge numbers and a correspondence in Quasitoric Orbifolds
Authors:
Saibal Ganguli
Abstract:
We give hodge structures on quasitoric orbifolds. We define orbifold hodge numbers and show a correspondence of orbifold hodge numbers for crepant resolutions of quasitoric orbifolds. In short we extend hodge structures to a non complex setting .
We give hodge structures on quasitoric orbifolds. We define orbifold hodge numbers and show a correspondence of orbifold hodge numbers for crepant resolutions of quasitoric orbifolds. In short we extend hodge structures to a non complex setting .
△ Less
Submitted 26 December, 2015; v1 submitted 18 June, 2014;
originally announced June 2014.
-
Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
Authors:
Yann Dauphin,
Razvan Pascanu,
Caglar Gulcehre,
Kyunghyun Cho,
Surya Ganguli,
Yoshua Bengio
Abstract:
A central challenge to many fields of science and engineering involves minimizing non-convex error functions over continuous, high dimensional spaces. Gradient descent or quasi-Newton methods are almost ubiquitously used to perform such minimizations, and it is often thought that a main source of difficulty for these local methods to find the global minimum is the proliferation of local minima wit…
▽ More
A central challenge to many fields of science and engineering involves minimizing non-convex error functions over continuous, high dimensional spaces. Gradient descent or quasi-Newton methods are almost ubiquitously used to perform such minimizations, and it is often thought that a main source of difficulty for these local methods to find the global minimum is the proliferation of local minima with much higher error than the global minimum. Here we argue, based on results from statistical physics, random matrix theory, neural network theory, and empirical evidence, that a deeper and more profound difficulty originates from the proliferation of saddle points, not local minima, especially in high dimensional problems of practical interest. Such saddle points are surrounded by high error plateaus that can dramatically slow down learning, and give the illusory impression of the existence of a local minimum. Motivated by these arguments, we propose a new approach to second-order optimization, the saddle-free Newton method, that can rapidly escape high dimensional saddle points, unlike gradient descent and quasi-Newton methods. We apply this algorithm to deep or recurrent neural network training, and provide numerical evidence for its superior optimization performance.
△ Less
Submitted 10 June, 2014;
originally announced June 2014.
-
Mckay Correspondence in Quasitoric Orbifolds
Authors:
Saibal Ganguli
Abstract:
We show Mckay correspondence of Betti numbers of Chen-Ruan coho- mology for omnioriented quasitoric orbifolds. In previous articles with M. Poddar [8], [9], we proved the correspondence for four dimension and six dimensions. Here we deal with the general case.
We show Mckay correspondence of Betti numbers of Chen-Ruan coho- mology for omnioriented quasitoric orbifolds. In previous articles with M. Poddar [8], [9], we proved the correspondence for four dimension and six dimensions. Here we deal with the general case.
△ Less
Submitted 19 August, 2013;
originally announced August 2013.
-
Almost complex structure, blowdowns and McKay correspondence in quasitoric orbifolds
Authors:
Saibal Ganguli,
Mainak Poddar
Abstract:
We prove the existence of invariant almost complex structure on any positively omnioriented quasitoric orbifold. We construct blowdowns. We define Chen-Ruan cohomology ring for any omnioriented quasitoric orbifold. We prove that the Euler characteristic of this cohomology is preserved by a crepant blowdown. We prove that the Betti numbers are also preserved if dimension is less or equal to six. In…
▽ More
We prove the existence of invariant almost complex structure on any positively omnioriented quasitoric orbifold. We construct blowdowns. We define Chen-Ruan cohomology ring for any omnioriented quasitoric orbifold. We prove that the Euler characteristic of this cohomology is preserved by a crepant blowdown. We prove that the Betti numbers are also preserved if dimension is less or equal to six. In particular, our work reveals a new form of McKay correspondence for orbifold toric varieties that are not Gorenstein. We illustrate with an example.
△ Less
Submitted 24 February, 2012;
originally announced February 2012.
-
Blowdowns and McKay correspondence on four dimensional quasitoric orbifolds
Authors:
Saibal Ganguli,
Mainak Poddar
Abstract:
We prove the existence of torus invariant almost complex structure on any positively omnioriented four dimensional primitive quasitoric orbifold. We construct pseudo-holomorphic blowdown maps for such orbifolds. We prove a version of McKay correspondence when the blowdowns are crepant.
We prove the existence of torus invariant almost complex structure on any positively omnioriented four dimensional primitive quasitoric orbifold. We construct pseudo-holomorphic blowdown maps for such orbifolds. We prove a version of McKay correspondence when the blowdowns are crepant.
△ Less
Submitted 15 October, 2011; v1 submitted 4 November, 2009;
originally announced November 2009.
