Our Universe is homogeneous and isotropic, and its perturbations obey translation and rotation symmetry. In this work, we develop translation and rotation equivariant normalizing flow (TRENF), a generative normalizing flow (NF) model which explicitly incorporates these symmetries, defining the data likelihood via a sequence of Fourier space-based convolutions and pixel-wise non-linear transforms. TRENF gives direct access to the high dimensional data likelihood p(x|y) as a function of the labels y, such as cosmological parameters. In contrast to traditional analyses based on summary statistics, the NF approach has no loss of information since it preserves the full dimensionality of the data. On Gaussian random fields, the TRENF likelihood agrees well with the analytical expression and saturates the Fisher information content in the labels y. On non-linear cosmological overdensity fields from N-body simulations, TRENF leads to significant improvements in constraining power over the standard power spectrum summary statistic. TRENF is also a generative model of the data, and we show that TRENF samples agree well with the N-body simulations it trained on, and that the inverse mapping of the data agrees well with a Gaussian white noise both visually and on various summary statistics: when this is perfectly achieved the resulting p(x|y) likelihood analysis becomes optimal. Finally, we develop a generalization of this model that can handle effects that break the symmetry of the data, such as the survey mask, which enables likelihood analysis on data without periodic boundaries.
- NSF-PAR ID:
- 10321738
- Date Published:
- Journal Name:
- Future Internet
- Volume:
- 13
- Issue:
- 7
- ISSN:
- 1999-5903
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
ABSTRACT -
Deconvolution is a key component in contemporary neural networks, especially generative adversarial networks (GANs) and fully convolutional networks (FCNs). Due to extra operations of deconvolution compared to convolution, considerable degradation of performance as well as energy efficiency is incurred when implementing deconvolution on the existing resistive random access memory (ReRAM)-based processing-in-memory (PIM) accelerators. In this work, we propose a ReRAM-based accelerator design, RED, for providing high-performance and low-energy deconvolution. We analyze the deconvolution execution on the existing ReRAM-based PIMs and utilize its interior computation pattern for design optimization. RED includes two major contributions: pixel-wise mapping scheme and zero-skipping data flow. Pixel-wise mapping scheme removes the zero insertion and performs convolutions over several ReRAM arrays and thus enables parallel computations with non-zero inputs. Zero-skipping data flow, assisted with customized input buffers design, enhances the computation parallelism and input data reuse. In evaluation, we compare RED against the existing ReRAM-based PIMs and CMOS-based counterpart with a variety of GAN and FCN models, each of which contains multiple deconvolution layers. The experimental results show that RED achieves a 4.0×-56.16× speedup and a 1.05×-18.17× energy efficiency improvement over previous related accelerator designs.more » « less
-
With reduced data reuse and parallelism, recent convolutional neural networks (CNNs) create new challenges for FPGA acceleration. Systolic arrays (SAs) are efficient, scalable architectures for convolutional layers, but without proper optimizations, their efficiency drops dramatically for reasons: 1) the different dimensions within same-type layers, 2) the different convolution layers especially transposed and dilated convolutions, and 3) CNN’s complex dataflow graph. Furthermore, significant overheads arise when integrating FPGAs into machine learning frameworks. Therefore, we present a flexible, composable architecture called FlexCNN, which delivers high computation efficiency by employing dynamic tiling, layer fusion, and data layout optimizations. Additionally, we implement a novel versatile SA to process normal, transposed, and dilated convolutions efficiently. FlexCNN also uses a fully-pipelined software-hardware integration that alleviates the software overheads. Moreover, with an automated compilation flow, FlexCNN takes a CNN in the ONNX representation, performs a design space exploration, and generates an FPGA accelerator. The framework is tested using three complex CNNs: OpenPose, U-Net, and E-Net. The architecture optimizations achieve 2.3 × performance improvement. Compared to a standard SA, the versatile SA achieves close-to-ideal speedups, with up to 15.98 × and 13.42 × for transposed and dilated convolutions, with a 6% average area overhead. The pipelined integration leads to a 5 × speedup for OpenPose.more » « less
-
more » « less
Abstract Transfer functions (i.e., convolutions) have been powerful tools for both forward modeling and reconstructing past states of transport systems in a wide range of subsurface flow problems. The advantage of the transfer function is that it is a simple alternative to complicated, distributed parameter models of flow and transport, but the majority of applications of transfer functions in hydrology have been limited to relatively simple cases, like passive tracers or first‐order decay. The central question evaluated in this note is whether or not multicomponent mixing‐limited reactive transport can be represented within a transfer function framework. Our examples consider forward‐in‐time predictions and backward‐in‐time reconstructions of a carbonate system that represents the intrusion of seawater into a freshwater aquifer. The main result is that accurate forward‐in‐time and backward‐in‐time models are developed by posing the problem in terms of conservative components. As with all convolution‐based methods, the results are sensitive to errors and/or noise in the input functions, but we show that smoothed approximations of the requisite functions provide good representations of transport. Given the vast unknowns in any subsurface transport problem, such generalized, reactive transfer function models may have yet unexplored advantages when the trade‐offs between overall computational cost, accuracy, and uncertainty are explored in more detail.
-
The unsupervised task of aligning two or more distributions in a shared latent space has many applications including fair representations, batch effect mitigation, and unsupervised domain adaptation. Existing flow-based approaches estimate multiple flows independently, which is equivalent to learning multiple full generative models. Other approaches require adversarial learning, which can be computationally expensive and challenging to optimize. Thus, we aim to jointly align multiple distributions while avoiding adversarial learning. Inspired by efficient alignment algorithms from optimal transport (OT) theory for univariate distributions, we develop a simple iterative method to build deep and expressive flows. Our method decouples each iteration into two subproblems: 1) form a variational approximation of a distribution divergence and 2) minimize this variational approximation via closed-form invertible alignment maps based on known OT results. Our empirical results give evidence that this iterative algorithm achieves competitive distribution alignment at low computational cost while being able to naturally handle more than two distributions.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/fi13070179
