The paper integrates AI and expert knowledge to optimize radiotherapy decisions.
problem Optimizing radiation dose planning considering patient-specific information.
method Integrating Gaussian process models with deep neural networks to quantify uncertainty.
result Improves AI model performance and guides clinical decision making.
A ML-based method reconstructs 3D organ doses from 2D radiographs for pediatric abdominal radiotherapy.
problem Reconstructing detailed 3D dose distributions for childhood cancer survivors using limited 2D radiographs.
method Surrogate-free ML approach using 142 abdominal planning CTs, 300 artificial plans, and evolutionary algorithm.
result Accurate 3D dose reconstructions with MAEs ≤ 1.7 Gy for edge organs, validated on independent dataset.
CNNs can be trained with clinically available segmentations for OARs in radiotherapy.
problem Lack of dedicated training volumes for CNNs in radiotherapy.
method Used clinically available segmentations from PACS, applied multi-label segmentation, empirically assessed training set size.
result Clinically available segmentations can be used to train an accurate OAR segmentation model.
Study improves machine learning for estimating survival treatment effects.
problem Estimating heterogeneous survival treatment effects in observational data.
method Flexible machine learning methods in the counterfactual framework, including AFT-BART-NP.
result AFT-BART-NP consistently yields best performance in terms of bias, precision, and frequentist coverage.
CAST models time-varying treatment effects in cancer patients.
problem Estimating treatment effects at fixed time points limits understanding of dynamic changes over time.
method CAST combines parametric and non-parametric methods to model continuous time-varying treatment effects.
result CAST reveals how treatment effects rise, peak, and decline over the follow-up period.
Over half a million individuals are diagnosed with head and neck cancer each year worldwide. Radiotherapy is an important curative treatment for this disease, but it requires manual time consuming delineation of radio-sensitive organs at risk (OARs). This planning process can delay treatment, while also introducing int…
Automates quality control for synthetic CTs generated from MR images.
problem Prevent downstream errors in RT treatment planning from synthetic CTs.
method Ensemble of sCT generators and uncertainty measure based on their disagreement.
result Uncertainty measure can detect input images outside expected MR distribution and sCT images with potential errors.
Study improves confidence measures in medical imaging pipelines by addressing bias.
problem Bias in metric-based imaging pipelines compromises the efficiency of prediction intervals.
method Formalized symmetric and asymmetric CP formulations, analyzed bias effects, and validated empirically.
result Symmetric intervals are inflated by bias, while asymmetric intervals remain unaffected.
Computed tomography (CT) equivalent information is needed for attenuation correction in PET imaging and for dose planning in radiotherapy. Prior work has shown that Gaussian mixture models can be used to generate a substitute CT (s-CT) image from a specific set of MRI modalities. This work introduces a more flexible cl…
Low precision weights, activations, and gradients have been proposed as a way to improve the computational efficiency and memory footprint of deep neural networks. Recently, low precision networks have even shown to be more robust to adversarial attacks. However, typical implementations of low precision DNNs use unifor…
Mixed-precision CA-SGD for generalized linear models on GPUs
problem SGD communication bottleneck
method Mixed-precision CA-SGD
result Matches FP32 SGD loss within 0.5% on various problems
Paper explores low-precision SGLD for neural networks, reducing costs without sacrificing performance.
problem Infeasibility of low-precision sampling in large-scale scenarios.
method Developed low-precision SGLD with quantization function and full-precision gradient accumulators.
result Low-precision SGLD achieves comparable performance to full-precision SGLD with only 8 bits.
Post-training quantization method using multiple low-precision points achieves higher precision for critical weights.
problem Discretizing pre-trained deep neural networks without re-training.
method Multipoint quantization with efficient greedy selection and adaptive point number.
result Outperforms state-of-the-art methods on ImageNet classification and PASCAL VOC object detection.
This paper improves low-precision sampling using SGHMC for deep learning models.
problem Enhancing training efficiency of deep neural networks with low-precision training.
method Investigates low-precision sampling via Stochastic Gradient Hamiltonian Monte Carlo (SGHMC) for both log-concave and non-log-concave distributions.
result Low-precision SGHMC achieves quadratic improvement in error compared to SGLD for non-log-concave distributions.
Low precision operations can provide scalability, memory savings, portability, and energy efficiency. This paper proposes SWALP, an approach to low precision training that averages low-precision SGD iterates with a modified learning rate schedule. SWALP is easy to implement and can match the performance of full-precisi…
Low-precision computation is often used to lower the time and energy cost of machine learning, and recently hardware accelerators have been developed to support it. Still, it has been used primarily for inference - not training. Previous low-precision training algorithms suffered from a fundamental tradeoff: as the num…
Efficient Bitwidth Search optimizes neural network quantization for better performance.
problem Finding optimal bitwidth for weights and activations of each layer efficiently.
method EBS algorithm reusing meta weights and binary decomposition for efficient mixed precision convolution.
result Mixed precision QNN outperforms uniform bitwidth and other techniques on CIFAR10 and ImageNet.
Deep neural networks have enabled progress in a wide variety of applications. Growing the size of the neural network typically results in improved accuracy. As model sizes grow, the memory and compute requirements for training these models also increases. We introduce a technique to train deep neural networks using hal…
The study examines how class imbalance affects precision-recall curves.
problem Understanding how precision changes with class imbalance ratios.
method Analyzes the relationship between precision, class imbalance ratio, and true/false positive rates.
result Predicts changes in precision-recall curves and other measures with class imbalance ratios.
It is well-known that the precision of data, hyperparameters, and internal representations employed in learning systems directly impacts its energy, throughput, and latency. The precision requirements for the training algorithm are also important for systems that learn on-the-fly. Prior work has shown that the data and…
This paper introduces a new method to train normalizing flows using precision-recall divergences.
problem Training generative models with mode dropping and low-quality samples.
method Introduces PR-divergences and proposes a novel generative model to minimize precision-recall trade-offs.
result Normalizing flows can be trained to achieve specific precision-recall trade-offs using PR-divergences.
A framework estimates multiple precision matrices with shared structures.
problem Estimating multiple precision matrices with shared structures.
method Penalized likelihood framework with iterative algorithm alternating between convex and clustering problems.
result The method outperforms competitors and performs similarly to methods using prior information.
Efforts to reduce the numerical precision of computations in deep learning training have yielded systems that aggressively quantize weights and activations, yet employ wide high-precision accumulators for partial sums in inner-product operations to preserve the quality of convergence. The absence of any framework to an…
Paper explores reducing precision in SVM for faster text classification.
problem Efficiency in multi-class text classification training.
method Comparison of SVM trained with reduced precision (16-bit, half) vs original.
result Reduced precision training maintains text classification accuracy.
Low-precision streaming PCA estimates the leading eigenvector with limited precision.
problem Estimating the leading eigenvector in a streaming setting with limited precision.
method Oja's algorithm with linear and nonlinear stochastic quantization.
result A batched version of the quantized variants achieves the lower bound on quantization error up to logarithmic factors.
Low-precision training reduces computational cost and produces efficient models. Recent research in developing new low-precision training algorithms often relies on simulation to empirically evaluate the statistical effects of quantization while avoiding the substantial overhead of building specific hardware. To suppor…
EMPIR combines low and full precision DNNs to enhance robustness against adversarial attacks.
problem Vulnerability of DNNs to adversarial attacks that misclassify inputs with small perturbations.
method Ensemble of quantized DNN models with different numerical precisions.
result EMPIR ensembles increase adversarial robustness by 42.6% on average across different tasks.
Lower bounds set for infinite-precision transformers.
problem Understanding limitations of infinite-precision transformers.
method Used VC dimension technique to prove lower bounds.
result First lower bounds for two tasks: function composition and SUM2. LLMs struggle with arithmetic tasks unless they use high numerical precision.
problem Improving arithmetical reasoning capabilities of LLMs.
method Theoretical analysis and empirical experiments on numerical precision.
result LLMs require high numerical precision to efficiently handle arithmetic tasks.
Precision farming uses data analysis to optimize crop management.
problem Meeting increased food demand with sustainable practices.
method Statistical tools and machine learning for pattern recognition and prediction.
result Optimized farming processes in real-time reduce resource use.
The paper improves Bayesian precision matrix estimation for high-dimensional sparse data.
problem Estimating sparse precision matrices in high-dimensional settings.
method Tempered posterior with fully specified horseshoe prior.
result Concentration results and theoretical oracle inequality for posterior.
Study precise asymptotics of noncompact Type-IIb solutions to mean curvature flow.
problem Understanding the behavior of noncompact Type-IIb solutions to mean curvature flow as time approaches infinity.
method Constructed rotationally symmetric solutions with specific asymptotic behavior and analyzed their properties.
result The highest curvature concentrates at the tip of the hypersurface and blows up at the Type-IIb rate (2t+1)(γ−1)/2. New proposed models are often compared to state-of-the-art using statistical significance testing. Literature is scarce for classifier comparison using metrics other than accuracy. We present a survey of statistical methods that can be used for classifier comparison using precision, accounting for inter-precision corre…
Reduced precision computation for deep neural networks is one of the key areas addressing the widening compute gap driven by an exponential growth in model size. In recent years, deep learning training has largely migrated to 16-bit precision, with significant gains in performance and energy efficiency. However, attemp…
FQ-Conv quantizes CNNs for efficient inference with low-precision weights and activations.
problem Reducing precision in DNNs leads to reduced accuracy.
method Fully quantized convolutional neural networks (FQ-Conv) using novel quantization and training techniques.
result Ternary-weight CNNs perform nearly as well as full-precision networks.
Gaussian graphical models (GGMs) are probabilistic tools of choice for analyzing conditional dependencies between variables in complex systems. Finding changepoints in the structural evolution of a GGM is therefore essential to detecting anomalies in the underlying system modeled by the GGM. In order to detect structur…
SINGD improves KFAC for memory-efficiency and stability in low-precision training.
problem Memory inefficiency and numerical instability of KFAC in low-precision training.
method Formulated inverse-free KFAC update and imposed structures in Kronecker factors.
result SINGD is memory-efficient and numerically robust, often outperforming AdamW in half precision.
A new R package for high-dimensional regression and precision matrix estimation.
problem High-dimensional linear regression and precision matrix estimation challenges.
method flare package implements various regression methods and extensions for sparse precision matrix estimation.
result The flare package is efficient and scalable for large problems.
Correct method found for drawing precise envelope of straight lines.
problem Widespread method fails to represent the precise shape of envelope.
method Recently discovered correct method for straight line families in the plane.
result Correct method precisely represents the envelope of straight lines.
Finite-precision learning of anh networks is limited by the Monte Carlo rate.
problem Learning anh neural networks under finite precision method Using iterated anh activations to construct localized bump functions result No adaptive randomized algorithm can achieve higher convergence rate than Monte Carlo rate in finite precision
Maximizing the speed and precision of communication while minimizing power dissipation is a fundamental engineering design goal. Also, biological systems achieve remarkable speed, precision and power efficiency using poorly understood physical design principles. Powerful theories like information theory and thermodynam…
HMQ improves quantization for edge devices with mixed precision.
problem Efficient quantization for edge devices with uniform, power-of-two thresholds.
method Introduces HMQ, a mixed precision quantization block that repurposes Gumbel-Softmax for searching over quantization schemes.
result Achieves competitive and state-of-the-art results on ImageNet despite restrictions.
Diagonal transformations preserve independence structures in non-Gaussian distributions.
problem Preserving independence structures in non-Gaussian distributions.
method Diagonal nonlinear transformations of multivariate normal variables.
result Independence structures are preserved in non-Gaussian distributions under diagonal transformations.
This work studies scaling laws for low-precision training in high-dimensional linear regression.
problem Optimizing trade-off between model quality and training costs in high-dimensional linear regression.
method Theoretical study of scaling laws for low-precision training within a high-dimensional sketched linear regression framework, analyzing multiplicative and additive quantization.
result Multiplicative quantization maintains full-precision model size, while additive quantization reduces effective model size.
Deep networks run with low precision operations at inference time offer power and space advantages over high precision alternatives, but need to overcome the challenge of maintaining high accuracy as precision decreases. Here, we present a method for training such networks, Learned Step Size Quantization, that achieves…
Precise computations of Dehn functions for subgroups of free group products.
problem Computing precise Dehn functions for subgroups of direct products of free groups.
method Analyzing specific subgroups and using algebraic methods to compute Dehn functions.
result Quartic and quadratic Dehn functions for specific subgroups of free group products.
The high computational and parameter complexity of neural networks makes their training very slow and difficult to deploy on energy and storage-constrained computing systems. Many network complexity reduction techniques have been proposed including fixed-point implementation. However, a systematic approach for designin…
With ever-increasing computational demand for deep learning, it is critical to investigate the implications of the numeric representation and precision of DNN model weights and activations on computational efficiency. In this work, we explore unconventional narrow-precision floating-point representations as it relates …