CompactNet optimizes CNN models for resource-limited platforms.
problem Challenges in implementing CNN models on resource-limited platforms.
method Guided by a simulator, CompactNet progressively trims a pre-trained network to achieve target speedup while maintaining accuracy.
result Achieves up to 1.8x kernel computation speedup on embedded platforms.
MoGA optimizes mobile networks for GPU efficiency.
problem Lack of GPU-aware optimization in mobile network design.
method Proposes MoGA, a GPU-aware neural architecture search.
result Achieves higher accuracy with similar latency constraints.
Automated framework optimizes DNN deployment on Arm CPUs.
problem Lack of globally optimised DNN deployment across software levels.
method Reinforcement Learning search for automated design space exploration.
result Up to 4x improvement in performance and 2x reduction in memory.
CUDA optimized neural network predicts HbA1c from joint mobility and anthropometrics.
problem Early detection and accurate diagnosis of diabetes.
method Parallelized neural network using CUDA and C++ on Nvidia GPUs.
result Achieved high accuracy (95.65% on training, 86.67% on testing for males; 97.73% on training, 66.67% on testing for females).
Mobile SoCs can improve ML inference by 2x when all components are used concurrently.
problem Power-performance trade-offs in mobile ML inference on heterogeneous SoCs.
method Quantitative evaluation of different ML components on mobile SoCs.
result Mobile SoCs can achieve up to 2x improvement in inference performance by using all components concurrently.
The paper optimizes neural network inference on mobile GPUs.
problem Limited computing power and thermal constraints on mobile CPUs.
method Leverage mobile GPUs for neural network inference.
result Real-time inference of deep neural networks on Android and iOS devices.
State of the art deep learning models have made steady progress in the fields of computer vision and natural language processing, at the expense of growing model sizes and computational complexity. Deploying these models on low power and mobile devices poses a challenge due to their limited compute capabilities and str…
A DRL approach optimizes resource allocation in BFL to reduce latency and energy consumption.
problem Energy and CPU constraints of mobile devices and increased training latency due to blockchain mining.
method Deep Reinforcement Learning (DRL) to derive optimal decisions for MLMO.
result Optimal resource allocation and block generation rate to minimize system latency, energy consumption, and incentive cost.
A new framework for mobile authentication using deep metric learning.
problem Challenges in mobile authentication using behavioral biometrics.
method Deep metric learning, private data protection, flexible training scheduling.
result 95% authentication accuracy on public datasets, robust against attacks.
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.
The paper tackles carousel personalization in music streaming apps using contextual bandits.
problem Selecting relevant items to display in carousels for personalized content recommendation.
method Modeling carousel personalization as a contextual multi-armed bandit problem with multiple plays, cascade-based updates and delayed batch feedback.
result Empirically shows the effectiveness of the framework in capturing characteristics of real-world carousels.
Computer vision performances have been significantly improved in recent years by Convolutional Neural Networks(CNN). Currently, applications using CNN algorithms are deployed mainly on general purpose hardwares, such as CPUs, GPUs or FPGAs. However, power consumption, speed, accuracy, memory footprint, and die size sho…
The rising popularity of intelligent mobile devices and the daunting computational cost of deep learning-based models call for efficient and accurate on-device inference schemes. We propose a quantization scheme that allows inference to be carried out using integer-only arithmetic, which can be implemented more efficie…
HRL4IN tackles interactive navigation tasks with mobile manipulators, improving efficiency and performance.
problem Interactive Navigation tasks require mobile manipulators to perform various actions, but choosing the right part of the embodiment is inefficient.
method HRL4IN uses a hierarchical reinforcement learning architecture to handle heterogeneous phases of navigation and manipulation, selecting the appropriate part of the embodiment for each phase.
result HRL4IN significantly outperforms flat PPO and HAC in terms of task performance and energy efficiency.
Convolutional Neural Networks (CNNs) are extremely computationally demanding, presenting a large barrier to their deployment on resource-constrained devices. Since such systems are where some of their most useful applications lie (e.g. obstacle detection for mobile robots, vision-based medical assistive technology), si…
New algorithm reduces regret in high-dimensional bandit problems.
problem High-dimensional bandit problems with sparse correlations.
method Doubly-Robust Lasso Bandit algorithm exploiting sparse regression structure.
result Regret bound scales with log(d) instead of polynomial of d.
Contextual multi-armed bandit (MAB) algorithms have been shown promising for maximizing cumulative rewards in sequential decision tasks such as news article recommendation systems, web page ad placement algorithms, and mobile health. However, most of the proposed contextual MAB algorithms assume linear relationships be…
We introduce a learning-based framework to optimize tensor programs for deep learning workloads. Efficient implementations of tensor operators, such as matrix multiplication and high dimensional convolution, are key enablers of effective deep learning systems. However, existing systems rely on manually optimized librar…
Performance-aware channel pruning improves CNN on embedded GPUs.
problem Inefficient channel pruning on embedded GPUs leads to performance slowdowns.
method Evaluate higher-level libraries that analyze input characteristics for optimized code generation.
result Performance-aware pruning can achieve significant performance speedups, up to 10x.
Ansor generates high-performance tensor programs for deep learning.
problem Generating high-performance tensor programs for deep learning on various hardware platforms is challenging.
method Ansor uses a hierarchical representation of the search space, sampling programs, and evolutionary search with a learned cost model to find high-performance programs.
result Ansor improves deep neural network execution performance up to 3.8x on Intel CPU, 2.6x on ARM CPU, and 1.7x on NVIDIA GPU.
Massively parallel architectures such as the GPU are becoming increasingly important due to the recent proliferation of data. In this paper, we propose a key class of hybrid parallel graphlet algorithms that leverages multiple CPUs and GPUs simultaneously for computing k-vertex induced subgraph statistics (called graph…
In the edge computing paradigm, mobile devices offload the computational tasks to an edge server by routing the required data over the wireless network. The full potential of edge computing becomes realized only if a smart device selects the most appropriate server in terms of the latency and energy consumption, among …
TASO optimizes CNN models for memory-constrained devices.
problem Large CNN models are too resource-intensive for mobile devices.
method Integer Linear Programming (ILP) for selecting operations and workspace allocation.
result 8x speedup and 2.2x reduction in memory usage compared to greedy algorithms.
New setting combines state evolution and corrupted context for better decision-making.
problem Decision-making in a changing state with unreliable context.
method Proposes a new algorithm using a referee to dynamically combine contextual bandit and multi-armed bandit policies.
result Improved empirical performance compared to existing algorithms.
LoRA fine-tuning on CPUs without GPUs achieves comparable performance to GPU-based methods.
problem Limited computational resources for fine-tuning LLMs.
method Developed a CPU-efficient method to learn meta-operators for LoRA weights.
result LoRA adapters trained on CPUs outperform base models on downstream tasks.
The paper optimizes RF training by improving tree building algorithms and CPU optimizations.
problem Improving the training performance of Random Forest models on CPU architectures.
method Investigated and improved tree building algorithms (BFS, DFS, hybrid BFS-DFS) and proposed optimizations.
result The hybrid BFS-DFS algorithm outperforms both BFS and DFS, and is more robust.
Deep learning solutions are being increasingly used in mobile applications. Although there are many open-source software tools for the development of deep learning solutions, there are no guidelines in one place in a unified manner for using these tools towards real-time deployment of these solutions on smartphones. Fr…
AI methods often fail to outperform classical CPU-based solvers on Maximum Independent Set problems.
problem Comparing AI methods with classical CPU-based solvers on Maximum Independent Set problems.
method Comparison of AI methods (e.g., generative models, reinforcement learning) with classical CPU-based solvers (e.g., KaMIS) on Maximum Independent Set problem.
result AI-inspired methods are often outperformed by classical CPU-based solvers, even with post-processing techniques.
Signatory calculates signature and logsignature transforms efficiently on CPU and GPU.
problem Efficient computation of signature and logsignature transforms for machine learning.
method CPU and GPU parallelism, backpropagation, efficient precomputation strategies, algorithmic improvements.
result Substantial speedups on CPU and GPU, including real-world applications.
Benchmarking TPU, GPU, and CPU for deep learning models.
problem Improving performance in deep learning training.
method ParaDnn benchmark suite for FC, CNN, and RNN models on TPU, GPU, and CPU.
result TPU, GPU, and CPU have unique strengths for different types of models.
We present Caffe con Troll (CcT), a fully compatible end-to-end version of the popular framework Caffe with rebuilt internals. We built CcT to examine the performance characteristics of training and deploying general-purpose convolutional neural networks across different hardware architectures. We find that, by employi…
CuLE accelerates reinforcement learning on GPUs.
problem Limited CPU-GPU communication bandwidth in Atari Learning Environment.
method CUDA port of ALE, leveraging GPU parallelization and direct frame rendering.
result Up to 155M frames per hour on a single GPU, accelerating reinforcement learning.
Asymmetry PRISM outperforms CPU and GPU solvers for institutional rebalancing.
problem Institutional rebalancing with deadline constraints
method Asymmetry PRISM
result Asymmetry PRISM-CPU is 4.5x to 24.1x faster than the fastest completed reference row in the same lane.
As machine learning algorithms enter applications in industrial settings, there is increased interest in controlling their cpu-time during testing. The cpu-time consists of the running time of the algorithm and the extraction time of the features. The latter can vary drastically when the feature set is diverse. In this…
Mobile sensing is an emerging technology that utilizes agent-participatory data for decision making or state estimation, including multimedia applications. This article investigates the structure of mobile sensing schemes and introduces crowdsourcing methods for mobile sensing. Inspired by social network, one can estab…
Simple method improves exploration in various decision problems.
problem Improving exploration in sequential decision problems.
method Parameterized Exploration (PE) method that considers time horizon and state of knowledge.
result PE outperforms un-tuned methods in various bandit and decision problem settings.
A decentralized policy achieves logarithmic regret for multi-agent MAB problems with communication constraints.
problem Decentralized policy for multi-agent MAB problems with option availability and communication constraints.
method Upper Confidence Bound (UCB) algorithms with non-stationary stochastic communication protocol.
result Guaranteed logarithmic regret for non-fully connected spatial graphs with communication constraints.
This is a brief survey of the research performed by Grandata Labs in collaboration with numerous academic groups around the world on the topic of human mobility. A driving theme in these projects is to use and improve Data Science techniques to understand mobility, as it can be observed through the lens of mobile phone…
Introduces mobility algebra for modeling geodesics on n-spheres.
problem Modeling geodesics on n-spheres using algebraic structures.
method Introduces mobility algebra and mobility spaces, showing connections to modules and affine spaces.
result Shows geodesics on n-spheres as mobility spaces over unit interval mobility algebra.
The paper presents a probabilistic method to discover daily human mobility patterns from mobile data.
problem Discovering daily human mobility patterns from mobile data.
method A non-parameter Bayesian modeling method, Infinite Gaussian Mixture Model, combined with Kullback-Leibler divergence for automatic clustering.
result The IGMM-based algorithm outperforms the GMM-based algorithm in discovering mobility patterns.
This study uses LSTM and SARIMA models to forecast CPU usage in cloud computing.
problem Predicting future CPU usage in cloud computing centers.
method Used LSTM and SARIMA models to forecast CPU usage over short and long time scales.
result SARIMA model performed better for long-term predictions but LSTM was more robust.
FLAME auto-labels mobile data efficiently on diverse processors.
problem Accurately and efficiently labeling mobile data with unknown labels on heterogeneous processors.
method Self-adaptive auto-labeling system Flame that schedules and executes workloads on mobile processors.
result Flame achieves high labeling accuracy and performance on heterogeneous mobile processors.
Graphs model human mobility patterns, reducing errors in data matching.
problem Lack of high-quality data and computational resources for graph-based mobility analysis.
method Embedding graphs into a continuous space to address matching, modeling, and visualization challenges.
result Approx 40% decrease in error on average in matched graphs vs unmatched ones.
pySigLib speeds up signature-based computations on CPUs and GPUs.
problem Efficient signature-based computations on large datasets and long sequences.
method Optimised Python library for CPU and GPU, novel differentiation scheme.
result Accurate gradients at a fraction of the runtime of existing libraries.
Model predicts urban population using mobile data traffic.
problem Estimating urban population dynamics from mobile data.
method Data-driven approach combining NetMob 2023 and ENACT datasets.
result NetMob 2023 data can estimate urban population with XGBoost models.
In this paper, we investigate the economic mobility in some money transfer models which have been applied into the research on wealth distribution. We demonstrate the mobility by recording the time series of agents' ranks and observing their volatility. We also compare the mobility quantitatively by employing an index,…
MobILE learns from expert demonstrations without access to actions, achieving strong performance guarantees.
problem Imitation Learning from Observations alone (ILFO) where actions are not observed.
method MobILE integrates optimism in the face of uncertainty into distribution matching IL framework, trading off exploration and imitation.
result MobILE provides strong performance guarantees for certain MDP dynamics and demonstrates efficacy on OpenAI Gym tasks.
New models predict mobility flows as well as complex machine learning but are simpler and interpretable.
problem Incomplete understanding and modeling of human mobility flows.
method Developed simple machine-learned, closed-form models of mobility.
result These models predict mobility flows more accurately than gravity or complex machine/deep learning models.