Approximate computing environments trade off computational accuracy for improvements in performance, energy, and resiliency cost. For widespread adoption of approximate computing, a fundamental requirement is to understand how perturbations to a computation affect the outcome of the execution in terms of its output quality. This paper presents a framework for approximate computing, called Approxilyzer, that quantifies the quality impact of a single-bit error in all dynamic instructions of an execution with high accuracy (95% on average). We demonstrate two uses of Approxilyzer.

SASSIFI is an architecture-level error injection-based methodology and tool to study the soft error resilience of massively parallel applications running on state-of-the-art NVIDIA GPUs. SASSIFI provides an automated flow to perform error injection campaigns and is publicly available on GitHub at https://github.com/NVlabs/sassifi.

We show that the equations of reinforcement learning and light transport simulation are related integral equations. Based on this correspondence, a scheme to learn importance while sampling path space is derived. The new approach is demonstrated in a consistent light transport simulation algorithm that uses reinforcement learning to progressively learn where light comes from.

While ray tracing has become increasingly common and path tracing is well understood by now, a major challenge lies in crafting an easy-to-use and efficient system implementing these technologies. Following a purely physically-based paradigm while still allowing for artistic workflows, the Iray light transport simulation and rendering system allows for rendering complex scenes by the push of a button and thus makes accurate light transport simulation widely available.

We present a GPU-based ray traversal algorithm that operates on compressed wide BVHs and maintains the traversal stack in a compressed format. Our method reduces the amount of memory traffic significantly, which translates to 1.9-2.1x improvement in incoherent ray traversal performance compared to the current state of the art. Furthermore, the memory consumption of our hierarchy is 35-60% of a typical uncompressed BVH.

Some of the fundamental limitations of existing near-eye displays (NEDs) for augmented reality (AR) are limited field of view (FOV), low angular resolution, and fixed focal state. Optimizing a design to overcome one of these limitations typically leads to a trade-off in the others. This project introduces an all-in-one solution: deformable beamsplitters. It is a new new hybrid hardware design created by combining hyperbolic half-silvered mirrors and deformable membrane mirrors (DMMs).

Accommodative depth cues, a wide field of view, and ever-higher resolutions all present major hardware design challenges for near-eye displays. Optimizing a design to overcome one of these challenges typically leads to a trade-off in the others. We tackle this problem by introducing an all-in-one solution - a new wide field of view, gaze-tracked near-eye display for augmented reality applications. The key component of our solution is the use of a single see-through, varifocal deformable membrane mirror for each eye reflecting a display.

In this paper we develop a parallel approach for computing the modularity clustering often used to identify and analyse communities in social networks. We show that modularity can be approximated by looking at the largest eigenpairs of the weighted graph adjacency matrix that has been perturbed by a rank one update. Also, we generalize this formulation to identify multiple clusters at once. We develop a fast parallel implementation for it that takes advantage of the Lanczos eigenvalue solver and k-means algorithm on the GPU.

The tools of soft robotics enable immersive kinesthetic experiences in virtual reality. Using fluidic elastomer actuators (FEAs), we demonstrate a soft skin that can provide force feedback and a soft controller to simulate different textures and materials. These novel input devices integrate with a VR Funhouse experience.