New algorithms improve learning deep energy models.
problem Learning deep energy models efficiently and accurately.
method Proposed new algorithms combining GAN-style methods with traditional energy-based learning.
result SteinCD performs well in test likelihood, SteinGAN in generating realistic images.
Proposes linking energy and force uncertainty in deep learning potentials.
problem Uncertainty in predicted energies and forces in machine learning models.
method Introduces a spatially correlated noise process to link energy and force uncertainty.
result Demonstrates the approach on molecular datasets, linking energy and force uncertainties.
Deep learning predicts energy loads and prices with LSTM and EVT.
problem Predicting extreme loads in energy grids due to supply and demand fluctuations.
method Deep spatio-temporal models and EVT for tail behavior of load spikes.
result Deep LSTM models outperform traditional methods in capturing nonlinearities.
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.
Enhanced tabular benchmarks for energy-efficient neural architecture search.
problem Energy consumption in deep learning models.
method Introducing EC-NAS, an enhanced tabular benchmark with energy consumption data.
result EC-NAS reveals a balance between energy usage and accuracy in neural architecture search.
HIP-NN models molecular energies using a deep neural network with hierarchical terms.
problem Accurately predicting molecular energies from quantum calculations.
method HIP-NN decomposes molecular properties into a sum of hierarchical terms generated by a neural network.
result Achieves state-of-the-art performance with 0.26 kcal/mol mean absolute error.
This review compares various deep generative models.
problem Training deep neural networks to model data distributions.
method Comprehensive comparison of VAEs, GANs, flows, energy models, and autoregressives.
result Trade-offs and interrelationships among different models.
Deep learning improves solar energy forecasting using physical and data-driven models.
problem Improving short-term solar energy forecasting accuracy.
method Injecting physical knowledge into deep learning models for spatio-temporal forecasting.
result Improved solar energy forecasting models using deep learning and physical criteria.
In this paper, we attack the anomaly detection problem by directly modeling the data distribution with deep architectures. We propose deep structured energy based models (DSEBMs), where the energy function is the output of a deterministic deep neural network with structure. We develop novel model architectures to integ…
Deep learning boosts building energy load forecasting.
problem Short-term load forecasting in buildings.
method Stacked Boosters Network architecture with sparse interactions, parameter sharing, and equivariant representations.
result Outperforms state-of-the-art models in short-term load forecasting tasks.
Scaling algorithms improve joint training of deep energy-based models.
problem Joint training of deep energy-based models often fails and delivers worse results.
method Proposed online and offline scaling algorithms to fix joint training.
result Scaling algorithms improve joint training and deliver better results.
The paper tackles energy disaggregation by improving dictionary learning with deep neural models.
problem Decomposing electricity signals of a whole home into its operating devices.
method Proposes a novel optimization program that learns both the dictionary and sparse coefficients using a deep neural model (LSTM-AE) to capture temporal energy signals.
result Significant improvement in disaggregation accuracy and F-score metrics compared to state-of-the-art methods.
Deep Autoencoder outperforms in anomaly detection for building energy data.
problem Automated detection of faulty data in learning applications.
method Training and comparison of Simple, Deep, and Supervised Deep Autoencoders on ASHRAE building energy dataset.
result Supervised Deep Autoencoder outperforms in total anomalies detected.
Deep reinforcement learning improves trading performance in volatile energy markets.
problem Volatility and low signal-to-noise ratios in energy markets.
method Formalized trading as a stochastic system, developed reactive and adaptive algorithms, used deep neural networks.
result Deep reinforcement learning models outperform buy-and-hold strategy with an 83% higher Sharpe ratio.
Paper optimizes neural network layers to reduce energy usage without sacrificing accuracy.
problem Energy consumption in deep neural networks during inference.
method Layerwise noise maximization to optimize reliability of memory elements.
result Reduces memory energy consumption by 3.3 times at equal accuracy.
Deep-Energy trains DNNs without labels using energy functions.
problem Training DNNs without manually annotated labels.
method Uses task-specific energy functions to train DNNs unsupervised.
result Trained DNNs provide better quality labels than direct minimization.
Deep learning depends on tuning layers near critical points.
problem Understanding how deep learning architectures depend on tuning parameters.
method Random energy approach to analyze statistical dependence in deep belief networks.
result Statistical dependence can propagate only if layers are tuned near critical points.
Training energy-based probabilistic models is confronted with apparently intractable sums, whose Monte Carlo estimation requires sampling from the estimated probability distribution in the inner loop of training. This can be approximately achieved by Markov chain Monte Carlo methods, but may still face a formidable obs…
Efficient neural network improves disaggregation of home energy usage.
problem Estimating power consumption of individual appliances from total home power.
method Fully convolutional neural network architecture with improved computational efficiency.
result Achieves state-of-the-art disaggregation performance with reduced training and prediction times.
ECC compresses DNNs for energy-constrained devices like UAVs and smartphones.
problem Energy-constrained deep neural networks in vision applications.
method ECC uses a bilinear regression model to estimate DNN energy consumption and optimizes compression to meet energy constraints.
result ECC achieves higher accuracy under the same or lower energy budget compared to state-of-the-art techniques.
Modeling wind dynamics in Saudi Arabia using deep learning and stochastic PDEs.
problem Accurately modeling spatio-temporal wind patterns in a large, diverse, and understudied region.
method Energy distance-based spatial reduction, sparse stochastic Echo State Network, non-stationary stochastic PDE reconstruction.
result Produces more accurate wind speed and energy forecasts, saving $1 million annually.
Paper tackles energy sharing in ZECs using DRL.
problem Improving energy status of ZECs through agent-based energy sharing.
method Modelled as a multi-agent environment, solved with DRL.
result Agents learn to collaborate and improve ZEC's energy status over time.
This paper tackles risk-aware energy scheduling for MEC networks with microgrids.
problem Risk in energy demand and supply for MEC networks powered by microgrids.
method Formulated an optimization problem with CVaR for energy consumption and generation, analyzed using a multi-agent stochastic game, derived solution with MADRL-based A3C algorithm.
result Significant performance gain by considering CVaR for high accuracy energy scheduling.
Unified detector calibration and simulation using MLE from generative models.
problem Combining detector calibration and simulation using traditional methods.
method Maximum likelihood estimation from conditional generative models.
result Prior-independent and non-Gaussian resolutions possible.
Energy-efficient DL inference for IoT devices reduces power consumption and improves performance.
problem Energy inefficiency in deep learning models for IoT devices.
method Energy-aware early exiting policy to balance energy consumption and inference accuracy.
result Accuracy and service rate improved up to 25% and 35% respectively.
Geometric Occam's Razor shapes deep learning solutions.
problem Understanding the regularization in over-parameterized neural networks.
method Analyzing the geometric model complexity and Dirichlet energy in neural networks.
result Over-parameterized neural networks are implicitly regularized by geometric model complexity.
Paper introduces normalizing flows for accurate probabilistic energy forecasting.
problem Uncertainty in renewable energy forecasting for power systems.
method Normalizing flows for direct learning of multivariate stochastic distributions.
result Normalizing flows outperform other deep learning models in probabilistic forecasting.
Deep k-Means compresses CNNs by clustering weights and re-training, reducing energy consumption.
problem High energy consumption and large parameter count in deep convolutions.
method Applying k-means clustering on convolutional layer weights, sharing K cluster centers, and re-training with hard assignments. result Significant reduction in energy consumption and compression ratio without accuracy loss.
Paper optimizes UAV navigation for IoT data freshness and energy efficiency.
problem Improving data freshness and connectivity for IoT devices with UAVs.
method Deep reinforcement learning model with experience replay for energy-efficient UAV trajectory optimization.
result The proposed approach is 3.6% and 3.13% more energy efficient than greedy and baseline methods.
A deep learning method solves nonlinear filtering problems efficiently.
problem Nonlinear filtering problem
method Deep splitting method combined with energy-based neural network approximation
result Computational efficiency and performance comparable to Kalman and bootstrap filters
Deep learning improves neutrino-nucleus interaction vertex reconstruction.
problem Vertex reconstruction of neutrino-nucleus interaction events.
method Combining energy and timing data for classification and regression tasks using deep learning.
result The model achieves 4.00% higher classification accuracy and 0.9919 higher regression accuracy than previous methods.
Deep neural networks are optimizable due to their multilayered structure.
problem Understanding why deep neural networks are easily optimizable despite their non-convex loss functions.
method Analysis of a spin glass model of deep neural networks using random matrix theory and algebraic geometry.
result The multilayered structure of deep neural networks leads to fewer stationary points, more clustered minima, and less severe tradeoffs between depth and width of minima.
Paper uses DRL for smart MG energy dispatch, improving stability and performance.
problem Improving energy dispatch in IoT-driven smart MGs with DGs, PVs, and batteries.
method Formulated POMDP model, proposed FH-DDPG and FH-RDPG algorithms, compared with baseline algorithms.
result Proposed algorithms enhance MG performance and stability under uncertainty.
This thesis tackles NILM challenges with a new dataset and efficient edge deployment techniques.
problem Limited datasets and high computational power for NILM deployment.
method Developed an interoperable data collection framework and introduced model compression techniques.
result Efficient edge deployment of NILM models for global scalability and sustainability.
Paper proposes f-EBM for training deep EBMs using various f-divergences.
problem Training deep EBMs with intractable partition functions.
method Introduces f-EBM framework and optimization algorithm for any f-divergence.
result f-EBM outperforms contrastive divergence and other f-divergences.
New deep learning model optimizes energy use in buildings.
problem Optimizing energy use and comfort in large buildings.
method Transformer-based metamodel trained with simulation and sensor data, calibrated with CMA-ES, optimized with multi-objective algorithms.
result Optimal settings reduce energy loads while maintaining thermal comfort and air quality.
Paper proposes a new method for estimating treatment effects using interpretable deep learning models.
problem Estimating treatment effects from observational data with interpretability.
method Proposes a novel objective function using energy distance balancing score and neural additive models for improved interpretability.
result Demonstrates superior performance over state-of-the-art methods in semi-synthetic experiments.
Paper proposes energy-efficient DNN training methods.
problem Energy-constrained deployment of deep neural networks.
method Weighted sparse projection and layer input masking integrated into DNN training.
result Framework provides higher accuracy with same or lower energy budgets.
SmartDeal reduces energy and storage costs for deep neural networks.
problem Heavy parameterization of deep neural networks leads to inefficient use of DRAM.
method SmartDeal decomposes weights into a small basis matrix and a structurally sparse coefficient matrix, quantized to power-of-2.
result Up to 2.44x energy efficiency improvement in inference and 10.56x reduction in training energy.
Dissipative SymODEN learns dynamics with dissipation and control from data.
problem Learning dynamics with dissipation and control from observed data.
method Dissipative SymODEN encodes port-Hamiltonian dynamics into a deep learning architecture.
result The learned model reveals key aspects of the system, such as inertia, dissipation, and potential energy.
Reweighting improves GAN accuracy without sacrificing statistical power.
problem Improving the fidelity of generative models.
method Post-hoc reweighting function applied to generated examples.
result Weighted GAN examples significantly improve accuracy.
A new deep learning method using Boolean logic reduces training and inference energy.
problem High computational and energy costs in deep learning training and inference.
method Introduces Boolean weights and inputs for efficient training using Boolean logic.
result Achieves full-precision accuracy in ImageNet classification and surpasses state-of-the-art results in semantic segmentation.
This paper reviews methods to create compact neural networks for IoT applications.
problem Complex deep neural networks are costly and slow, hindering real-world deployment.
method Automatic synthesis of compact, accurate DNN/LSTM models.
result Compact neural networks reduce energy consumption, memory, and inference time.
The paper proposes a dataflow-based approach to reduce energy consumption in deep neural networks without sacrificing performance.
problem Reducing energy consumption in deep neural networks without performance drop.
method Dataflow-based joint quantization of weights and activations.
result Joint quantization improves performance and reduces energy consumption.
Optimizes deep reinforcement learning for energy-efficient video streaming.
problem Minimizing energy consumption in video streaming over mobile networks.
method Integrates DDPG algorithm with partially known model to reduce signaling overhead and improve convergence speed.
result Proposed policy converges to optimal policy with improved convergence speed.
This paper examines how energy in feature maps decays in deep neural networks.
problem Understanding energy decay in deep convolutional neural networks.
method Analyzes energy conservation and decay rates in various deep neural network architectures.
result Energy in feature maps decays polynomially or exponentially across layers.
Deep learning models predict solar irradiance for short-term forecasts.
problem Accurate prediction of solar irradiance for renewable energy integration.
method Sequence-to-sequence LSTM models for GHI forecasting, incorporating spatial-temporal features.
result LSTM models outperform traditional techniques in short-term GHI forecasting.
Paper uses deep learning to suppress bones on chest X-rays.
problem Improving pathologies classification by suppressing bones on chest X-rays.
method Conditional Generative Adversarial Network (GAN) and Haar 2D wavelet decomposition.
result Achieves state-of-the-art performance on bone suppression.