Dynamic programming (DP) is a widely used algorithmic paradigm, particularly in bioinformatics, finding applications in a wide spectrum of tasks, including read assembly, homology search, gene annotation, basecalling, and phylogenetic inference. Due to its computationally intensive nature, many ASIC- and FPGA-based accelerators have been proposed in recent years to accelerate specific tasks. However, DP algorithms in bioinformatics can vary considerably, and most existing solutions are customized for a single application, representing just one design point within the broader DP space. These implementations typically rely on low-level hardware description languages (HDLs), often requiring months of manual implementation effort. This paper introduces DP-HLS, a novel framework based on High-Level Synthesis (HLS) that simplifies and accelerates the development of a vast set of bioinformatically relevant 2-D DP algorithms in hardware. DP-HLS achieves this by introducing a new abstraction layer that decouples the front-end specification from predefined HLS-based back-end optimizations, enabling users to efficiently develop new 2-D DP kernels in C++ and deploy them on FPGAs without needing any expertise in hardware design or HLS. In our experience, DP-HLS significantly reduced the development time of new kernels (months to days) and produced designs with comparable resource utilization to open-source hand-coded HDL-based implementations and performance within$\mathbf{7. 7 - 1 6. 8 \%}$margin. DP-HLS is compatible with AWS®EC2 F1 FPGA instances. To showcase its versatility, we implemented 15 diverse 2-D DP kernels using the DP-HLS framework, achieving$1.38-41 \times$improved cost-efficiency over state-of-the-art GPU and CPU baselines and providing the first open-source FPGA implementation for several of them. The DP-HLS codebase is available freely under the MIT license at https://github.com/TurakhiaLab/DP-HLS.

The growing complexity in manufacturing and operating the hardware in AI clusters leads to significant challenges in reliability. Hyperscalars have reported various AI hardware failures during high-stake jobs such as GenAI model training, where one GPU failure could bring down the entire training job. To tackle this issue, we present CP-Bench, an open-source, Configurable and Parameterizable, PyTorch-level test suite designed to test AI hardware failure, performance degradation, and silent data corruption (SDC). Built upon open-source projects, CP-Bench contains 30+ AI workloads (e.g., Llama), and implements various checks (e.g., SDC check) within these workloads. We have deployed CP-Bench throughout Meta’s AI hardware lifecycle, spanning manufacturing, in-production diagnostics, and device RMA; CP-Bench identified various hardware issues, some of which were not caught by vendor’s tooling. Notably, vendor has acknowledged to establish CP-Bench as a valid RMA criteria and plan to integrate CP-Bench into its tooling. CP-Bench is open-sourced at https://github.com/facebookincubator/CP-Bench.

This paper introduces two extensions to the popular PyTorch machine learning framework, TorchDynamo and TorchInductor, which implement the torch.compile feature released in PyTorch 2. TorchDynamo is a Python-level just-in-time (JIT) compiler that enables graph compilation in PyTorch programs without sacrificing the flexibility of Python. It achieves this by dynamically modifying Python bytecode before execution and extracting sequences of PyTorch operations into an FX graph, which is then JIT compiled using one of many extensible backends. TorchInductor is the default compiler backend for TorchDynamo, which translates PyTorch programs into OpenAI's Triton for GPUs and C++ for CPUs. Results show that TorchDynamo is able to capture graphs more robustly than prior approaches while adding minimal overhead, and TorchInductor is able to provide a 2.27× inference and 1.41× training geometric mean speedup on an NVIDIA A100 GPU across 180+ real-world models, which outperforms six other compilers. These extensions provide a new way to apply optimizations through compilers in eager mode frameworks like PyTorch.

Explanation techniques that synthesize small, interpretable changes to a given image while producing desired changes in the model prediction have become popular for introspecting black-box models. Commonly referred to as counterfactuals, the synthesized explanations are required to contain discernible changes (for easy interpretability) while also being realistic (consistency to the data manifold). In this paper, we focus on the case where we have access only to the trained deep classifier and not the actual training data. While the problem of inverting deep models to synthesize images from the training distribution has been explored, our goal is to develop a deep inversion approach to generate counterfactual explanations for a given query image. Despite their effectiveness in conditional image synthesis, we show that existing deep inversion methods are insufficient for producing meaningful counterfactuals. We propose DISC (Deep Inversion for Synthesizing Counterfactuals) that improves upon deep inversion by utilizing (a) stronger image priors, (b) incorporating a novel manifold consistency objective and (c) adopting a progressive optimization strategy. We find that, in addition to producing visually meaningful explanations, the counterfactuals from DISC are effective at learning classifier decision boundaries and are robust to unknown test-time corruptions.

The UT Austin Villa team, from the University of Texas at Austin, won the 2015 RoboCup 3D Simulation League, winning all 19 games that the team played. During the course of the competition the team scored 87 goals and conceded only 1. Additionally the team won the RoboCup 3D Simulation League technical challenge by winning each of a series of three league challenges: drop-in player, kick accuracy, and free challenge. This paper describes the changes and improvements made to the team between 2014 and 2015 that allowed it to win both the main competition and each of the league technical challenges. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.