FedLog reduces communication in federated learning by sharing data summaries.
problem Significant communication overhead in federated learning with large model parameters.
method Shares minimal sufficient statistics via Bayesian inference and differential privacy.
result High learning accuracy with low communication overhead.
SOTERIA optimizes neural networks for secure inference with minimal overhead.
problem Protecting user privacy in ML-as-a-service models with low overhead.
method Neural architecture search with dual objectives of accuracy and cryptographic efficiency.
result SOTERIA constructs efficient models for secure inference.
Generative Adapter adapts LMs with a single forward pass, reducing inference overhead.
problem Efficient adaptation of large language models for new contexts.
method Generative Adapter directly maps new contexts to low-rank LM adapters via self-supervised learning.
result Significant reduction in inference overhead with no need for fine-tuning.
This paper optimizes AI inference on edge devices with reduced communication and computation costs.
problem Efficiently performing AI inference on resource-constrained edge devices with reduced communication and computation costs.
method A three-step framework for effective inference: model split point selection, communication-aware model compression, and task-oriented encoding of intermediate features.
result Our proposed framework achieves a better trade-off and significantly reduces inference latency compared to baseline methods.
New method reduces fine-tuning cost for reused models.
problem Repeating fine-tuning costs with outdated foundation models.
method Portable Reward Tuning (PRT) trains a reward model to maximize the same loss function as fine-tuning.
result PRT achieves comparable accuracy to inference-time tuning with less inference cost.
BayesAdapter turns pre-trained NNs into reliable BNNs with minimal overhead.
problem Scalability, accessibility, and reliability of Bayesian neural networks.
method Bayesian fine-tuning of pre-trained deterministic NNs to variational BNNs.
result BayesAdapter produces more reliable posteriors with less training overhead.
A new algorithm improves Bayesian federated learning by reducing communication overhead.
problem Bayesian federated learning constraints, including privacy, data ownership, and communication overhead.
method Proposes Quantised Langevin Stochastic Dynamics (QLSD) for Bayesian federated learning, using gradient compression and variance reduction techniques.
result Non-asymptotic and asymptotic convergence guarantees for QLSD and its improved versions.
We introduce CheckNet, a method for secure inference with deep neural networks on untrusted devices. CheckNet is like a checksum for neural network inference: it verifies the integrity of the inference computation performed by untrusted devices to 1) ensure the inference has actually been performed, and 2) ensure the i…
Paper proposes efficient BNN inference flow to reduce computation and memory costs.
problem High computation complexity in Bayesian Neural Networks (BNNs) limits deployment in power-constrained systems.
method Feature decomposition and memorization strategy to reduce computations and a memory-friendly computing framework to reduce memory overhead.
result Reduces computation by about half and energy consumption by 73% with 14% area overhead.
We present a sampling-free approach for computing the epistemic uncertainty of a neural network. Epistemic uncertainty is an important quantity for the deployment of deep neural networks in safety-critical applications, since it represents how much one can trust predictions on new data. Recently promising works were pr…
Efficient FPGA virtualization for deep learning reduces user isolation and overhead.
problem Poor isolation and heavy re-compilation overhead in FPGA-based DNN accelerators.
method Two-level instruction dispatch module, multi-core hardware resources pool, tiling-based instruction frame package, two-stage static-dynamic compilation.
result 1.07-1.69x and 1.88-3.12x throughput improvement over previous designs.
We propose Edward, a Turing-complete probabilistic programming language. Edward defines two compositional representations---random variables and inference. By treating inference as a first class citizen, on a par with modeling, we show that probabilistic programming can be as flexible and computationally efficient as t…
We introduce overdispersed black-box variational inference, a method to reduce the variance of the Monte Carlo estimator of the gradient in black-box variational inference. Instead of taking samples from the variational distribution, we use importance sampling to take samples from an overdispersed distribution in the s…
Deep Gaussian Processes improve likelihood-free inference for complex distributions.
problem Limited flexibility of Bayesian Optimization with GPs for multimodal distributions.
method Proposes Deep Gaussian Processes (DGPs) as a surrogate model for likelihood-free inference.
result DGPs outperform GPs on multimodal distributions while maintaining comparable performance on unimodal cases.
A major obstacle for widespread deployment of frequency division duplex (FDD)-based Massive multiple-input multiple-output (MIMO) communications is the large signaling overhead for reporting full downlink (DL) channel state information (CSI) back to the basestation (BS), in order to enable closed-loop precoding. We com…
DarkneTZ protects edge devices from DNN model leaks using TEE and model partitioning.
problem Privacy risks of pre-trained DNNs on edge devices through membership inference attacks.
method Model partitioning into sensitive and untrusted parts, leveraging TEE.
result DarkneTZ provides reliable model privacy with minimal performance overhead.
Adaptive learning method identifies and corrects corrupted data.
problem Robust learning from corrupted training sets.
method Identifies corrupted and non-corrupted samples with latent Bernoulli variables, formulates as likelihood maximization with marginalized latent variables, solved via variational inference and Expectation-Maximization.
result Improves over state-of-the-art by automatically inferring corruption level with minimal overhead.
The impact of the maximally possible batch size (for the better runtime) on performance of graphic processing units (GPU) and tensor processing units (TPU) during training and inference phases is investigated. The numerous runs of the selected deep neural network (DNN) were performed on the standard MNIST and Fashion-M…
Monte-Carlo Tree Search (MCTS) methods are drawing great interest after yielding breakthrough results in computer Go. This paper proposes a Bayesian approach to MCTS that is inspired by distributionfree approaches such as UCT [13], yet significantly differs in important respects. The Bayesian framework allows potential…
DriftLite improves inference quality of diffusion models without retraining.
problem Adapting pre-trained diffusion models to new target distributions without retraining.
method Lightweight, training-free particle-based approach that steers inference dynamics with optimal stability control.
result Consistently reduces variance and improves sample quality over existing methods.
New method calibrates neural SBI to avoid overconfident posteriors.
problem Overconfident posteriors in SBI due to inaccurate uncertainty quantification.
method Introduces a calibration term into neural model training objective, enabling end-to-end backpropagation.
result Achieves competitive or better coverage and posterior density than existing methods.
New method calibrates neural network uncertainty for medical images.
problem Uncalibrated probabilistic outputs from deep neural networks in medical diagnosis.
method Functional space variational inference for Bayesian neural networks.
result Better calibrated uncertainty estimates at lower computational cost.
BNRE improves simulation-based inference by producing more conservative posteriors.
problem Overconfident posteriors from current simulation-based inference algorithms risk false inferences.
method Balanced Neural Ratio Estimation (BNRE) that produces more conservative posterior approximations.
result BNRE produces more conservative posterior surrogates on all tested benchmarks and simulation budgets.
A basic question in the theory of fault-tolerant quantum computation is to understand the fundamental resource costs for performing a universal logical set of gates on encoded qubits to arbitrary accuracy. Here we consider qubits encoded with constant space overhead (i.e. finite encoding rate) in the limit of arbitrari…
Proposes efficient sensitivity analysis for complex Bayesian models.
problem Inefficiency of sensitivity analyses in complex Bayesian models.
method SA-ABI: weight sharing and neural network rapid inference.
result Efficiently integrates sensitivity analyses into Bayesian inference.
Paper proposes an AutoML framework for efficient device-edge co-inference.
problem Finding optimal hyper-parameters for model sparsity and feature compression.
method Sequential decision problem solved using deep reinforcement learning (DRL).
result Achieves better communication-computation trade-off and significant speedup.
New method detects if data points were used in training models with low cost and high power.
problem Detecting if a particular data point was used in training a model.
method Fine-grained modeling of null hypothesis in likelihood ratio tests, leveraging reference models and population data.
result RMIA has superior test power compared to prior methods, even at extremely low false positive rates.
Efficient algorithm finds fast Transformer models.
problem Slow inference time of Transformer models.
method Decompose Transformer architecture into components, use sampling-based one-shot search.
result Achieved 10% to 30% speedup on pre-trained BERT and 70% on top of a previous state-of-the-art model.
Deep learning reduces training overhead in massive MIMO systems.
problem Reducing training overhead in massive MIMO systems.
method Use of deep learning (NNs) to improve CSI acquisition and feedback processes.
result Significant improvements in performance and reduced complexity.
We consider the problem of learning a binary classifier from n different data sources, among which at most an η fraction are adversarial. The overhead is defined as the ratio between the sample complexity of learning in this setting and that of learning the same hypothesis class on a single data distribution. We pr…
A new method for graph neural networks speeds up inference and training.
problem Challenges in constructing mini-batches for large graphs in graph neural networks.
method Theoretical model of batch construction via maximizing influence score of nodes on outputs.
result Accelerates inference by up to 130x compared to previous methods.
SSNL improves simulation-based inference for high-dimensional data.
problem Performance degradation in neural likelihood estimation for high-dimensional data.
method Surjective Sequential Neural Likelihood (SSNL) using surjective normalizing flow models.
result SSNL avoids manual crafting of summary statistics and outperforms state-of-the-art methods.
FAQ efficiently evaluates LLMs with statistical guarantees using adaptive query selection.
problem Efficiently evaluating many LLMs on a large suite of benchmarks is expensive.
method FAQ uses Bayesian factor models, adaptive sampling, and proactive active inference to select queries.
result FAQ delivers up to 5x effective sample size gains over baselines, matching CI width with fewer queries.
Hypersolvers enable fast continuous-depth models for practical applications.
problem Infinite-depth models like Neural ODEs are computationally infeasible for large problems.
method Introducing hypersolvers, neural networks that solve ODEs efficiently with theoretical guarantees.
result Hypersolvers achieve comparable inference time to traditional discrete networks, making continuous-depth models practical.
VMoER improves uncertainty quantification in MoE layers for scalable foundation models.
problem Uncertainty quantification in large-scale models like MoE layers.
method Structured Bayesian approach with amortized variational inference over routing logits and temperature parameter inference.
result Improves routing stability, reduces calibration error, and increases AUROC by 12%.
Proposes a new method to approximate Gaussian inference in classification tasks.
problem Uncertainty quantification in classification tasks using softmax functions.
method Develops a new formalism to approximate Gaussian distributions over logit space and proposes using element-wise normCDF or sigmoid instead of softmax.
result Improves uncertainty quantification compared to softmax Monte Carlo sampling.
Convolutional Neural Networks (CNN) are being actively explored for safety-critical applications such as autonomous vehicles and aerospace, where it is essential to ensure the reliability of inference results in the presence of possible memory faults. Traditional methods such as error correction codes (ECC) and Triple …
We study the collaborative PAC learning problem recently proposed in Blum et al.~\cite{BHPQ17}, in which we have k players and they want to learn a target function collaboratively, such that the learned function approximates the target function well on all players' distributions simultaneously. The quality of the col…
I2C enables agents to learn efficient communication without redundancy.
problem Redundant broadcast communication in multi-agent cooperation.
method I2C learns a prior for agent-agent communication via causal inference and reinforcement learning.
result I2C reduces communication overhead and improves multi-agent cooperative performance.
AMF-VI uses adaptive mixtures of flows for robust VI across diverse distributions.
problem Inconsistent behavior of single-flow models across different distributions.
method Sequential expert training of individual flows and adaptive global weight estimation via likelihood-driven updates.
result AMF-VI achieves lower negative log-likelihood and stable gains in transport metrics across various posterior families.
Improved inference efficiency for complex simulations.
problem Challenges in performing inference under resource-intensive stochastic simulators.
method Active sequential neural posterior estimation (ASNPE) integrating active learning into posterior estimation.
result Improved sample efficiency with low computational overhead.
URGE improves diffusion model quality without gradients or Hessian.
problem Improving sample quality in diffusion models without gradient evaluations.
method Path-wise importance reweighting via Girsanov change of measure.
result URGE achieves better generation quality than existing methods.
We propose a framework for the derivation and evaluation of distributed iterative algorithms for receiver cooperation in interference-limited wireless systems. Our approach views the processing within and collaboration between receivers as the solution to an inference problem in the probabilistic model of the whole sys…
Noise injection improves inference privacy in DNN models.
problem Malicious servers can infer sensitive attributes from input data.
method Adaptive Noise Injection (ANI) using a lightweight DNN on the client.
result Significant improvement in privacy (up to 48.5% degradation in sensitive-task accuracy with <1% degradation in primary accuracy).
A framework to explain decoder-only sequence classification models using intermediate predictions.
problem Explaining predictions of decoder-only sequence classification models.
method Progressive Inference framework with Single Pass-Progressive Inference and Multi Pass-Progressive Inference methods.
result Significantly better attributions compared to prior work on text classification tasks.
Bayesian inference of discrete component states in civil infrastructures using PGMs and GNNs.
problem Inferring discrete states of civil infrastructure components from measurable responses is an ill-posed inverse problem.
method The study proposes a novel Bayesian inversion paradigm based on Probabilistic Graphical Models (PGMs) and Graph Neural Networks (GNNs). PGMs are used to model the problem, with parameters learned from data and structural topology prior. Inference is accomplished by GNNs, and a graph property-based training strategy is developed.
result The proposed framework effectively solves the challenges of inferring the posterior PDF for discrete variables in high-dimensional problems.
A new quantization strategy reduces Transformer model size and inference time.
problem Heavy computation load and memory overhead in Transformer models for mobile devices.
method Mixed precision quantization with varying bits per word in embedding blocks.
result 11.8x smaller model size and 3.5x speed up for on-device NMT.
PolyLUT uses polynomials to reduce FPGA latency.
problem Reducing latency in FPGA-based neural network inference.
method Training neural networks using multivariate polynomials as basic building blocks.
result Achieved significant latency and area improvements.