Casimir Wierzynski

Noah Shutty, Casimir Wierzynski

Recent work has constructed neural networks that are equivariant to continuous symmetry groups such as 2D and 3D rotations. This is accomplished using explicit Lie group representations to derive the equivariant kernels and nonlinearities. We present three contributions motivated by frontier applications of equivariance beyond rotations and translations. First, we relax the requirement for explicit Lie group representations with a novel algorithm that finds representations of arbitrary Lie groups given only the structure constants of the associated Lie algebra. Second, we provide a self-contained method and software for building Lie group-equivariant neural networks using these representations. Third, we contribute a novel benchmark dataset for classifying objects from relativistic point clouds, and apply our methods to construct the first object-tracking model equivariant to the Poincaré group.

Peitong Duan, Casimir Wierzynski, Lama Nachman

Automating parts of the user interface (UI) design process has been a longstanding challenge. We present an automated technique for optimizing the layouts of mobile UIs. Our method uses gradient descent on a neural network model of task performance with respect to the model's inputs to make layout modifications that result in improved predicted error rates and task completion times. We start by extending prior work on neural network based performance prediction to 2-dimensional mobile UIs with an expanded interaction space. We then apply our method to two UIs, including one that the model had not been trained on, to discover layout alternatives with significantly improved predicted performance. Finally, we confirm these predictions experimentally, showing improvements up to 9.2 percent in the optimized layouts. This demonstrates the algorithm's efficacy in improving the task performance of a layout, and its ability to generalize and improve layouts of new interfaces.

Michael Y. -S. Fang, Sasikanth Manipatruni, Casimir Wierzynski et al.

For the benefit of designing scalable, fault resistant optical neural networks (ONNs), we investigate the effects architectural designs have on the ONNs' robustness to imprecise components. We train two ONNs -- one with a more tunable design (GridNet) and one with better fault tolerance (FFTNet) -- to classify handwritten digits. When simulated without any imperfections, GridNet yields a better accuracy (~98%) than FFTNet (~95%). However, under a small amount of error in their photonic components, the more fault tolerant FFTNet overtakes GridNet. We further provide thorough quantitative and qualitative analyses of ONNs' sensitivity to varying levels and types of imprecisions. Our results offer guidelines for the principled design of fault-tolerant ONNs as well as a foundation for further research.

Fabian Boemer, Anamaria Costache, Rosario Cammarota et al.

In previous work, Boemer et al. introduced nGraph-HE, an extension to the Intel nGraph deep learning (DL) compiler, that enables data scientists to deploy models with popular frameworks such as TensorFlow and PyTorch with minimal code changes. However, the class of supported models was limited to relatively shallow networks with polynomial activations. Here, we introduce nGraph-HE2, which extends nGraph-HE to enable privacy-preserving inference on standard, pre-trained models using their native activation functions and number fields (typically real numbers). The proposed framework leverages the CKKS scheme, whose support for real numbers is friendly to data science, and a client-aided model using a two-party approach to compute activation functions. We first present CKKS-specific optimizations, enabling a 3x-88x runtime speedup for scalar encoding, and doubling the throughput through a novel use of CKKS plaintext packing into complex numbers. Second, we optimize ciphertext-plaintext addition and multiplication, yielding 2.6x-4.2x runtime speedup. Third, we exploit two graph-level optimizations: lazy rescaling and depth-aware encoding, which allow us to significantly improve performance. Together, these optimizations enable state-of-the-art throughput of 1,998 images/s on the CryptoNets network. Using the client-aided model, we also present homomorphic evaluation of (to our knowledge) the largest network to date, namely, pre-trained MobileNetV2 models on the ImageNet dataset, with 60.4\percent/82.7\percent\ top-1/top-5 accuracy and an amortized runtime of 381 ms/image.

Fabian Boemer, Yixing Lao, Rosario Cammarota et al.

Homomorphic encryption (HE)---the ability to perform computation on encrypted data---is an attractive remedy to increasing concerns about data privacy in deep learning (DL). However, building DL models that operate on ciphertext is currently labor-intensive and requires simultaneous expertise in DL, cryptography, and software engineering. DL frameworks and recent advances in graph compilers have greatly accelerated the training and deployment of DL models to various computing platforms. We introduce nGraph-HE, an extension of nGraph, Intel's DL graph compiler, which enables deployment of trained models with popular frameworks such as TensorFlow while simply treating HE as another hardware target. Our graph-compiler approach enables HE-aware optimizations-- implemented at compile-time, such as constant folding and HE-SIMD packing, and at run-time, such as special value plaintext bypass. Furthermore, nGraph-HE integrates with DL frameworks such as TensorFlow, enabling data scientists to benchmark DL models with minimal overhead.