Improvements & Evaluations on the MLCommons CloudMask Benchmark

In this paper, we report the performance benchmarking results of deep learning models on MLCommons' Science cloud-masking benchmark using a high-performance computing cluster at New York University (NYU): NYU Greene. MLCommons is a consortium that develops and maintains several scientific benchmarks that can benefit from developments in AI. We provide a description of the cloud-masking benchmark task, updated code, and the best model for this benchmark when using our selected hyperparameter settings. Our benchmarking results include the highest accuracy achieved on the NYU system as well as the average time taken for both training and inference on the benchmark across several runs/seeds. Our code can be found on GitHub. MLCommons team has been kept informed about our progress and may use the developed code for their future work.

Sustainable Supercomputing for AI: GPU Power Capping at HPC Scale

As research and deployment of AI grows, the computational burden to support and sustain its progress inevitably does too. To train or fine-tune state-of-the-art models in NLP, computer vision, etc., some form of AI hardware acceleration is virtually a requirement. Recent large language models require considerable resources to train and deploy, resulting in significant energy usage, potential carbon emissions, and massive demand for GPUs and other hardware accelerators. However, this surge carries large implications for energy sustainability at the HPC/datacenter level. In this paper, we study the aggregate effect of power-capping GPUs on GPU temperature and power draw at a research supercomputing center. With the right amount of power-capping, we show significant decreases in both temperature and power draw, reducing power consumption and potentially improving hardware life-span with minimal impact on job performance. While power-capping reduces power draw by design, the aggregate system-wide effect on overall energy consumption is less clear; for instance, if users notice job performance degradation from GPU power-caps, they may request additional GPU-jobs to compensate, negating any energy savings or even worsening energy consumption. To our knowledge, our work is the first to conduct and make available a detailed analysis of the effects of GPU power-capping at the supercomputing scale. We hope our work will inspire HPCs/datacenters to further explore, evaluate, and communicate the impact of power-capping AI hardware accelerators for more sustainable AI.

Quantized Side Tuning: Fast and Memory-Efficient Tuning of Quantized Large Language Models

Finetuning large language models (LLMs) has been empirically effective on a variety of downstream tasks. Existing approaches to finetuning an LLM either focus on parameter-efficient finetuning, which only updates a small number of trainable parameters, or attempt to reduce the memory footprint during the training phase of the finetuning. Typically, the memory footprint during finetuning stems from three contributors: model weights, optimizer states, and intermediate activations. However, existing works still require considerable memory and none can simultaneously mitigate memory footprint for all three sources. In this paper, we present Quantized Side Tuing (QST), which enables memory-efficient and fast finetuning of LLMs by operating through a dual-stage process. First, QST quantizes an LLM's model weights into 4-bit to reduce the memory footprint of the LLM's original weights; QST also introduces a side network separated from the LLM, which utilizes the hidden states of the LLM to make task-specific predictions. Using a separate side network avoids performing backpropagation through the LLM, thus reducing the memory requirement of the intermediate activations. Furthermore, QST leverages several low-rank adaptors and gradient-free downsample modules to significantly reduce the trainable parameters, so as to save the memory footprint of the optimizer states. Experiments show that QST can reduce the total memory footprint by up to 2.3 $\times$ and speed up the finetuning process by up to 3 $\times$ while achieving competent performance compared with the state-of-the-art. When it comes to full finetuning, QST can reduce the total memory footprint up to 7 $\times$.

Modeling multi-legged robot locomotion with slipping and its experimental validation

Multi-legged robots with six or more legs are not in common use, despite designs with superior stability, maneuverability, and a low number of actuators being available for over 20 years. This may be in part due to the difficulty in modeling multi-legged motion with slipping and producing reliable predictions of body velocity. Here we present a detailed measurement of the foot contact forces in a hexapedal robot with multiple sliding contacts, and provide an algorithm for predicting these contact forces and the body velocity. The algorithm relies on the recently published observation that even while slipping, multi-legged robots are principally kinematic, and employ a friction law ansatz that allows us to compute the shape-change to body-velocity connection and the foot contact forces. This results in the ability to simulate motion plans for a large number of potentially slipping legs. In homogeneous environments, this can run in (parallel) logarithmic time of the planning horizon

From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference

Large language models (LLMs) have exploded in popularity due to their new generative capabilities that go far beyond prior state-of-the-art. These technologies are increasingly being leveraged in various domains such as law, finance, and medicine. However, these models carry significant computational challenges, especially the compute and energy costs required for inference. Inference energy costs already receive less attention than the energy costs of training LLMs -- despite how often these large models are called on to conduct inference in reality (e.g., ChatGPT). As these state-of-the-art LLMs see increasing usage and deployment in various domains, a better understanding of their resource utilization is crucial for cost-savings, scaling performance, efficient hardware usage, and optimal inference strategies. In this paper, we describe experiments conducted to study the computational and energy utilization of inference with LLMs. We benchmark and conduct a preliminary analysis of the inference performance and inference energy costs of different sizes of LLaMA -- a recent state-of-the-art LLM -- developed by Meta AI on two generations of popular GPUs (NVIDIA V100 \& A100) and two datasets (Alpaca and GSM8K) to reflect the diverse set of tasks/benchmarks for LLMs in research and practice. We present the results of multi-node, multi-GPU inference using model sharding across up to 32 GPUs. To our knowledge, our work is the one of the first to study LLM inference performance from the perspective of computational and energy resources at this scale.

Combining Explicit and Implicit Regularization for Efficient Learning in Deep Networks

Works on implicit regularization have studied gradient trajectories during the optimization process to explain why deep networks favor certain kinds of solutions over others. In deep linear networks, it has been shown that gradient descent implicitly regularizes toward low-rank solutions on matrix completion/factorization tasks. Adding depth not only improves performance on these tasks but also acts as an accelerative pre-conditioning that further enhances this bias towards low-rankedness. Inspired by this, we propose an explicit penalty to mirror this implicit bias which only takes effect with certain adaptive gradient optimizers (e.g. Adam). This combination can enable a degenerate single-layer network to achieve low-rank approximations with generalization error comparable to deep linear networks, making depth no longer necessary for learning. The single-layer network also performs competitively or out-performs various approaches for matrix completion over a range of parameter and data regimes despite its simplicity. Together with an optimizer's inductive bias, our findings suggest that explicit regularization can play a role in designing different, desirable forms of regularization and that a more nuanced understanding of this interplay may be necessary.

A Heterogeneous Parallel Non-von Neumann Architecture System for Accurate and Efficient Machine Learning Molecular Dynamics

This paper proposes a special-purpose system to achieve high-accuracy and high-efficiency machine learning (ML) molecular dynamics (MD) calculations. The system consists of field programmable gate array (FPGA) and application specific integrated circuit (ASIC) working in heterogeneous parallelization. To be specific, a multiplication-less neural network (NN) is deployed on the non-von Neumann (NvN)-based ASIC (SilTerra 180 nm process) to evaluate atomic forces, which is the most computationally expensive part of MD. All other calculations of MD are done using FPGA (Xilinx XC7Z100). It is shown that, to achieve similar-level accuracy, the proposed NvN-based system based on low-end fabrication technologies (180 nm) is 1.6x faster and 10^2-10^3x more energy efficiency than state-of-the-art vN based MLMD using graphics processing units (GPUs) based on much more advanced technologies (12 nm), indicating superiority of the proposed NvN-based heterogeneous parallel architecture.

Energy-Efficient Blockchain-enabled User-Centric Mobile Edge Computing

In the traditional mobile edge computing (MEC) system, the availability of MEC services is greatly limited for the edge users of the cell due to serious signal attenuation and inter-cell interference. User-centric MEC (UC-MEC) can be seen as a promising solution to address this issue. In UC-MEC, each user is served by a dedicated access point (AP) cluster enabled with MEC capability instead of a single MEC server, however, at the expense of more energy consumption and greater privacy risks. To achieve efficient and reliable resource utilization with user-centric services, we propose an energy efficient blockchain-enabled UC-MEC system where blockchain operations and resource optimization are jointly performed. Firstly, we design a resource-aware, reliable, replicated, redundant, and fault-tolerant (R-RAFT) consensus mechanism to implement secure and reliable resource trading. Then, an optimization framework based on alternating direction method of multipliers (ADMM) is proposed to minimize the total energy consumed by wireless transmission, consensus and task computing, where APs clustering, computing resource allocation and bandwidth allocation are jointly considered. Simulation results show superiority of the proposed UC-MEC system over reference schemes, at most 33.96% reduction in the total delay and 48.77% reduction in the total energy consumption.

A Green(er) World for A.I

As research and practice in artificial intelligence (A.I.) grow in leaps and bounds, the resources necessary to sustain and support their operations also grow at an increasing pace. While innovations and applications from A.I. have brought significant advances, from applications to vision and natural language to improvements to fields like medical imaging and materials engineering, their costs should not be neglected. As we embrace a world with ever-increasing amounts of data as well as research and development of A.I. applications, we are sure to face an ever-mounting energy footprint to sustain these computational budgets, data storage needs, and more. But, is this sustainable and, more importantly, what kind of setting is best positioned to nurture such sustainable A.I. in both research and practice? In this paper, we outline our outlook for Green A.I. -- a more sustainable, energy-efficient and energy-aware ecosystem for developing A.I. across the research, computing, and practitioner communities alike -- and the steps required to arrive there. We present a bird's eye view of various areas for potential changes and improvements from the ground floor of AI's operational and hardware optimizations for datacenters/HPCs to the current incentive structures in the world of A.I. research and practice, and more. We hope these points will spur further discussion, and action, on some of these issues and their potential solutions.

AI Enabled Maneuver Identification via the Maneuver Identification Challenge

Artificial intelligence (AI) has enormous potential to improve Air Force pilot training by providing actionable feedback to pilot trainees on the quality of their maneuvers and enabling instructor-less flying familiarization for early-stage trainees in low-cost simulators. Historically, AI challenges consisting of data, problem descriptions, and example code have been critical to fueling AI breakthroughs. The Department of the Air Force-Massachusetts Institute of Technology AI Accelerator (DAF-MIT AI Accelerator) developed such an AI challenge using real-world Air Force flight simulator data. The Maneuver ID challenge assembled thousands of virtual reality simulator flight recordings collected by actual Air Force student pilots at Pilot Training Next (PTN). This dataset has been publicly released at Maneuver-ID.mit.edu and represents the first of its kind public release of USAF flight training data. Using this dataset, we have applied a variety of AI methods to separate "good" vs "bad" simulator data and categorize and characterize maneuvers. These data, algorithms, and software are being released as baselines of model performance for others to build upon to enable the AI ecosystem for flight simulator training.