New KWS neural networks improve accuracy and power efficiency.
problem Maximizing accuracy and power efficiency in KWS systems.
method Hardware Aware Training (HAT) for LMU-based neural networks.
result Achieved state-of-the-art accuracy and low parameter counts.
Recent breakthroughs in Deep Learning (DL) applications have made DL models a key component in almost every modern computing system. The increased popularity of DL applications deployed on a wide-spectrum of platforms have resulted in a plethora of design challenges related to the constraints introduced by the hardware…
New activation networks improve model efficiency and performance.
problem Creating hardware-efficient deep learning models.
method Restructurable Activation Networks (RANs) with RAN-explicit and RAN-implicit methods.
result RANs achieve state-of-the-art results with improved hardware efficiency.
FrostNet improves INT8 quantization efficiency in mobile networks.
problem The importance of network architecture for optimal INT8 quantization.
method Quantization-aware training (QAT) with StatAssist and GradBoost, hardware-aware NAS.
result FrostNets achieve higher recognition accuracy with comparable latency when quantized.
Quantum algorithm reduces CVA risk-neutral expectation estimation costs.
problem Reducing Monte Carlo sampling cost for CVA on real quantum hardware.
method Noise-aware quantum workflow combining market calibration, discretisation, and oracle construction.
result CABIQAE achieves lower classical post-processing runtime and more effective error exploitation.
We implement a differentiable Neural Architecture Search (NAS) method inspired by FBNet for discovering neural networks that are heavily optimized for a particular target device. The FBNet NAS method discovers a neural network from a given search space by optimizing over a loss function which accounts for accuracy and …
Optimized neural networks for Edge TPU achieve high accuracy in real-time image classification.
problem Designing neural networks for hardware accelerators to achieve optimal performance.
method Hardware-aware neural architecture search and model customization for Edge TPU.
result Improved accuracy-latency tradeoff on Pixel 4's Edge TPU compared to existing models.
APQ jointly optimizes neural architecture, pruning, and quantization for efficient inference.
problem Efficient deep learning inference on resource-constrained hardware.
method Joint optimization of neural architecture, pruning, and quantization policy using a quantization-aware accuracy predictor.
result Joint optimization leads to 2.3% higher ImageNet accuracy with reduced latency and energy consumption.
We present an overview of techniques for quantizing convolutional neural networks for inference with integer weights and activations. Per-channel quantization of weights and per-layer quantization of activations to 8-bits of precision post-training produces classification accuracies within 2% of floating point networks…
OpEvo automates tensor operator optimization for better efficiency.
problem Manual optimization of tensor operators is inefficient and limited.
method OpEvo uses evolutionary computation with topology-aware mutation.
result OpEvo finds optimal configurations with less effort and variance.
DONNA rapidly finds optimal neural networks across diverse spaces.
problem Efficient scaling and handling of diverse architectural search-spaces in NAS.
method Three-phase pipeline: accuracy predictor, rapid evolutionary search, and optimal model finetuning.
result 100x faster than MNasNet in finding state-of-the-art architectures on-device.
NASCaps automates CapsNet design for better accuracy and hardware efficiency.
problem Designing Capsule Networks is laborious and inefficient.
method Automated Neural Architecture Search (NAS) with Genetic Algorithm optimization.
result Jointly optimizes network accuracy and hardware efficiency.
StatQAT optimizes quantization for deep networks, reducing computational cost and memory usage.
problem Optimal quantization parameters selection for deep neural networks with diverse data distributions.
method Statistical error analysis framework for uniform and floating-point quantization, iterative and analytic quantizers designed for arbitrary and Gaussian-like distributions.
result Improved accuracy and stability in training low-precision neural networks.
In recent years, Convolutional Neural Network (CNN) based methods have achieved great success in a large number of applications and have been among the most powerful and widely used techniques in computer vision. However, CNN-based methods are computational-intensive and resource-consuming, and thus are hard to be inte…
EDCompress optimizes energy efficiency of CNN models on edge devices.
problem Low energy consumption for edge devices with diverse dataflow types.
method Energy-aware model compression using reinforcement learning.
result Improves energy efficiency by 20X, 17X, 37X in various networks.
We formalize the problem of trading-off DNN training time and memory requirements as the tensor rematerialization optimization problem, a generalization of prior checkpointing strategies. We introduce Checkmate, a system that solves for optimal rematerialization schedules in reasonable times (under an hour) using off-t…
The paper addresses challenges in edge deep learning for IoT, proposing new directions.
problem Challenges in large-scale deep learning adoption for IoT devices.
method Unified view targeting three research directions: federated learning, data-independent deployment, and communication-aware inference.
result A network-centric approach is needed for edge intelligence.
PCNN prunes CNN weights efficiently for hardware acceleration.
problem Efficiently compressing CNN models for hardware acceleration.
method PCNN uses a novel Sparsity Pattern Mask (SPM) to encode and prune weights.
result PCNN achieves up to 8.4X compression with minimal accuracy loss.
Quantum state preparation framework speeds up basket option pricing.
problem Limited practical benefit of quantum amplitude estimation due to state-preparation depth.
method Structure-aware tensor-train rank-based variational state preparation.
result State-preparation depth scaling replaced with linear scaling, maintaining low basket-pricing errors.
Convolutional neural networks (CNNs) demand huge DRAM bandwidth for computational imaging tasks, and block-based processing has recently been applied to greatly reduce the bandwidth. However, the induced additional computation for feature recomputing or the large SRAM for feature reusing will degrade the performance or…
Pipelined Backpropagation trains large models without batches efficiently.
problem Training large models efficiently on hardware with limited batch sizes.
method Fine-grained Pipelined Backpropagation with Spike Compensation and Linear Weight Prediction.
result Fine-grained Pipelined Backpropagation with a batch size of one matches the accuracy of SGD for multiple networks.
Carbontracker tracks and predicts training DL models' carbon footprint.
problem Exponential growth in energy consumption for training deep learning models.
method Carbontracker tool for tracking and predicting energy and carbon footprint.
result Promotes responsible computing and encourages energy-efficient deep learning.
NEST optimizes deep learning training by placing devices efficiently across networks and memory.
problem Inefficient device placement in distributed deep learning leads to high communication and memory overhead.
method NEST uses network-, compute-, and memory-aware dynamic programming to optimize device placement.
result NEST achieves up to 2.43 times higher throughput and better memory efficiency.
Regularizes decision trees to reduce inference time by up to 4x with minimal accuracy loss.
problem Optimizing decision tree execution time on resource-constrained devices.
method Regularizes impurity computation during CART algorithm training to favor highly asymmetric distributions.
result Reduces inference time by up to 4x with minimal accuracy loss.
We use GG distributions to optimize LLMs, reducing size and training time.
problem Lack of understanding in the statistical structure of LLMs.
method GG-based initialization, ACT, GCT.
result Smaller, faster models with minimal communication overhead.
The significant computational requirements of deep learning present a major bottleneck for its large-scale adoption on hardware-constrained IoT-devices. Here, we envision a new paradigm called EdgeAI to address major impediments associated with deploying deep networks at the edge. Specifically, we discuss the existing …
Improves VQAs by balancing classical and quantum training resources.
problem Challenges in trainability and resource costs of VQAs on quantum hardware.
method Adopting HELIA Ansatz and combining classical and quantum methods for gradient estimation and training.
result Achieves higher accuracy and success rates in VQE and improved test accuracy in quantum phase classification.
Hardware-accelerated RBM solves large combinatorial problems and integer factorization.
problem Solving large combinatorial optimization and integer factorization problems.
method Logically synthesized RBM architecture, hardware acceleration, and efficient training methods.
result Hardware-accelerated RBM factorizes 16-bit numbers with 10000x speed and 32x power improvements.
Bio-inspired neuromorphic hardware is a research direction to approach brain's computational power and energy efficiency. Spiking neural networks (SNN) encode information as sparsely distributed spike trains and employ spike-timing-dependent plasticity (STDP) mechanism for learning. Existing hardware implementations of…
Resource-efficient oblique trees reduce neural signal classification costs.
problem Implementing efficient neural signal classifiers on resource-constrained devices.
method Integrating model compression, probabilistic routing, and cost-aware learning.
result Significant reduction in model size and feature extraction cost compared to state-of-the-art models.
QTAML models quantum tunneling errors for AI robustness.
problem Quantum tunneling errors in AI inference.
method Derives weight-error distribution using WKB approximation, introduces TAC algorithm.
result TAC achieves 95% clean accuracy with 3.4-33.6x less ECC overhead.
Paper develops a statistical framework for quantized training of deep neural networks.
problem Lack of theoretical understanding of gradient quantization in FQT.
method Presented a statistical framework for analyzing FQT algorithms, viewing quantized gradient as a stochastic estimator of QAT gradient.
result Developed two novel gradient quantizers with smaller variance than existing per-tensor quantizer.
Generative model for HW-aware DNN quantization.
problem Efficiently tuning DNNs for various hardware platforms.
method Generative model AQGAN for generating quantization configurations based on target accuracy.
result Generative model generates quantization configurations with competitive accuracy and lower search cost.
Improved neural network training with ADMM for hardware compatibility.
problem Training scalable and parallelizable neural networks.
method Alternating Direction Method of Multipliers (ADMM) combined with iterative least-square methods.
result Achieved significant accuracy improvements over SGD and Adam.
Quantum annealing (QA) is a hardware-based heuristic optimization and sampling method applicable to discrete undirected graphical models. While similar to simulated annealing, QA relies on quantum, rather than thermal, effects to explore complex search spaces. For many classes of problems, QA is known to offer computat…
This work reveals symmetries in quantum circuits and develops a noise-aware optimization method.
problem Understanding and optimizing the cost landscape of parametrized quantum circuits.
method Analytical proof of symmetries and their resilience to noise, followed by the development of SYMH optimization method.
result Symmetries in PQCs lead to degeneracy in the cost landscape and can be exploited to improve optimization under noise.
Photonic co-processor speeds up training of large neural networks.
problem Training large neural networks with backpropagation is inefficient and communication is a bottleneck.
method Direct Feedback Alignment (DFA) with a photonic accelerator.
result Photonic accelerator can compute random projections with trillions of parameters.
The intrinsic error tolerance of neural network (NN) makes approximate computing a promising technique to improve the energy efficiency of NN inference. Conventional approximate computing focuses on balancing the efficiency-accuracy trade-off for existing pre-trained networks, which can lead to suboptimal solutions. In…
This dissertation advances scalable Gaussian processes using iterative methods and pathwise conditioning.
problem The classical Gaussian process formulation is not scalable for large datasets and modern hardware.
method Combining iterative methods and pathwise conditioning to improve scalability.
result Significantly reduced memory requirements and facilitated application to larger datasets.
On-device CNN inference for real-time computer vision applications can result in computational demands that far exceed the energy budgets of mobile devices. This paper proposes FixyNN, a co-designed hardware accelerator platform which splits a CNN model into two parts: a set of layers that are fixed in the hardware pla…
Recent machine learning methods use increasingly large deep neural networks to achieve state of the art results in various tasks. The gains in performance come at the cost of a substantial increase in computation and storage requirements. This makes real-time implementations on limited resources hardware a challenging …
This paper highlights new opportunities for designing large-scale machine learning systems as a consequence of blurring traditional boundaries that have allowed algorithm designers and application-level practitioners to stay -- for the most part -- oblivious to the details of the underlying hardware-level implementatio…
This paper proposes a hardware-oriented dropout algorithm, which is efficient for field programmable gate array (FPGA) implementation. In deep neural networks (DNNs), overfitting occurs when networks are overtrained and adapt too well to training data. Consequently, they fail in predicting unseen data used as test data…
Graph neural networks have become increasingly popular in recent years due to their ability to naturally encode relational input data and their ability to scale to large graphs by operating on a sparse representation of graph adjacency matrices. As we look to scale up these models using custom hardware, a natural assum…
New method uses surrogate gradients to train efficient spiking networks on neuromorphic hardware.
problem Training high-performing spiking networks on analog neuromorphic hardware is challenging due to device mismatch and lack of efficient algorithms.
method Introduces a general in-the-loop learning framework based on surrogate gradients.
result Learning self-corrects for device mismatch, resulting in competitive spiking network performance.
Quantization can improve the execution latency and energy efficiency of neural networks on both commodity GPUs and specialized accelerators. The majority of existing literature focuses on training quantized DNNs, while this work examines the less-studied topic of quantizing a floating-point model without (re)training. …
Real-time semantic segmentation for autonomous vehicles on FPGA reduces latency and power consumption.
problem Efficient real-time semantic segmentation for autonomous vehicles.
method Compressed ENet architecture, FPGA deployment, batch processing, filter reduction, quantization-aware training.
result Reduced latency to 3 ms per image with batch size of ten and 40% resource utilization.
HotNAS reduces AI search time from hundreds of GPU hours to less than 3 GPU hours.
problem High time required for AI solution generation from scratch.
method HotNAS starts from a 'hot' state using pre-trained models and integrates compression during co-search.
result Reduces search time from 200 GPU hours to less than 3 GPU hours.