TOPNet integrates task-based evaluation into machine learning models.
problem Non-differentiable task-based evaluation criteria in real-world applications.
method Task-Oriented Prediction Network (TOPNet) with learnable surrogate loss function.
result TOPNet significantly outperforms traditional and heuristic models in financial prediction tasks.
New reward function improves GAIL performance in task-based environments.
problem Reward bias in adversarial imitation learning.
method Proposed a new reward function to overcome existing biases.
result New reward function outperforms existing methods in task-based environments.
Paper tackles catastrophic forgetting with task-based hard attention.
problem Catastrophic forgetting in neural networks after learning new tasks.
method Task-based hard attention mechanism learned through SGD.
result Reduces catastrophic forgetting by 45-80%.
ADRL improves participant selection in MCS systems.
problem Designing a participant selection algorithm for different MCS systems with multiple goals.
method Auxiliary-task based deep reinforcement learning (ADRL) using transformers and pointer networks.
result ADRL outperforms other baselines in various MCS settings.
Hybrid methods that utilize both content and rating information are commonly used in many recommender systems. However, most of them use either handcrafted features or the bag-of-words representation as a surrogate for the content information but they are neither effective nor natural enough. To address this problem, w…
CasVAE outperforms supervised methods for star-galaxy classification.
problem Challenges in machine learning for astronomy data.
method Cascade Variational Auto-Encoder (CasVAE) for unsupervised star-galaxy classification.
result CasVAE outperforms baseline models in accuracy and stability.
Framework adapts to new tasks based on prior knowledge.
problem Models struggle to adapt to novel tasks without direct experience.
method Learned task representations and meta-mappings to transform them.
result Meta-mapping achieves 80-90% performance on novel tasks.
We present a two-stage approach for learning dictionaries for object classification tasks based on the principle of information maximization. The proposed method seeks a dictionary that is compact, discriminative, and generative. In the first stage, dictionary atoms are selected from an initial dictionary by maximizing…
The ability to learn tasks in a sequential fashion is crucial to the development of artificial intelligence. Neural networks are not, in general, capable of this and it has been widely thought that catastrophic forgetting is an inevitable feature of connectionist models. We show that it is possible to overcome this lim…
Efficient privacy-preserving machine learning framework using random transformations.
problem Slow training and inference speed in privacy-preserving machine learning systems.
method Random transformations like linear and permutation, combined with arithmetic sharing.
result High efficiency and low computation cost in private machine learning.
A new framework uses information theory to detect anomalies in images without labeled data.
problem Detect anomalies in images without labeled data.
method A direct objective function using information theory to maximize the distance between normal and anomalous data.
result The proposed framework significantly outperforms state-of-the-arts on multiple benchmark datasets.
This research uses machine learning to approximate ideal and hotelling observer performance for binary signal detection.
problem Approximating the Ideal and Hotelling Observers for binary signal detection tasks.
method Supervised learning methods, including CNNs and SLNNs, are employed to approximate the IO and HO test statistics.
result The proposed supervised learning methods provide accurate approximations of the IO and HO test statistics.
Deep task-based quantization improves MIMO signal processing.
problem Improving performance of MIMO signal processing with scalar ADCs.
method Data-driven task-oriented quantization using deep learning.
result Deep task-based quantization can approach optimal performance limits.
Deep-RLS uses deep learning to improve PCA for better source separation.
problem Improving PCA for better source separation in nonlinear systems.
method Inspired by RLS, Deep-RLS unfolds RLS iterations into a deep neural network.
result Deep-RLS significantly improves accuracy in recovering source signals.
2L-FUSE enhances feature sparsity through kernel learning.
problem Sparsity and feature selection in regression tasks.
method 2-Layered kernel machines for learning a shape matrix and feature direction identification.
result Minimal yet informative feature sets are identified without losing predictive performance.
A new method uses PDEs to predict spatiotemporal phenomena.
problem Predicting high-dimensional spatiotemporal data.
method Partial differential equations (PDEs) for spatiotemporal disentanglement.
result The method outperforms existing models in accuracy and applicability.
This study uses CNN-IOs to estimate MRI image reconstruction performance bounds.
problem Estimating task-based performance limits for MRI image reconstruction methods.
method Utilized stylized multi-coil SENSE MRI systems and deep-generated stochastic models to estimate IO performance.
result Estimation of IO performance provides guidance for designing under-sampled MRI systems.
Detects student engagement states using unobtrusive appearance, context, and mouse data.
problem Detecting student engagement states in a classroom setting.
method Multimodal approach combining appearance, context, and mouse data; classifiers are fused at the decision level.
result Effective detection of student engagement states in a classroom setting.
Improves NMT with user feedback from eBay ratings and search tasks.
problem Improving neural machine translation quality with user feedback.
method Offline bandit learning of NMT parameters using real user feedback from eBay.
result Implicit task-based feedback from cross-lingual search tasks improves NMT quality.
Enhanced rotation prediction improves SSL models by capturing both shape and texture information.
problem Rotation prediction misses texture information, limiting model performance.
method Introduces image enhanced rotation prediction (IE-Rot) that combines rotation and image enhancement tasks.
result IE-Rot models outperform Rotation on various benchmarks.
Paper introduces CRLMaze, a new benchmark for continual reinforcement learning in 3D non-stationary environments.
problem Challenges of training reinforcement learning agents in high-dimensional, always-changing environments.
method End-to-end model-free continual reinforcement learning strategy.
result Competitive results in a complex 3D non-stationary task, outperforming four baselines.
One of the main challenges of deep learning methods is the choice of an appropriate training strategy. In particular, additional steps, such as unsupervised pre-training, have been shown to greatly improve the performances of deep structures. In this article, we propose an extra training step, called post-training, whi…
A new method for learning network representations that avoids information bias and sparsity.
problem Information bias and sparsity in network representation learning.
method A spreading-activation schema for learning node embeddings in network structures.
result Significant improvement in various real-world network analysis tasks.
Deep learning faces challenges in real-world tasks.
problem Challenges in applying deep learning to novel tasks without existing baselines.
method Case studies from research & development in conjunction with industry.
result Best practices for deep learning in practice.
This work disentangles speech and non-speech components from found data.
problem Building robust acoustic models from found data with non-standard variations.
method Latent Stochastic Models and Multinode Latent Space Variational Autoencoders (VAE).
result Speech and music can be separated in the latent space of a VAE, independent of the language.
MELEE learns good exploration strategies for contextual bandits.
problem Interactive contextual bandit exploration trade-off.
method Meta-learning from synthetic data to learn a good exploration policy.
result MELEE outperforms seven strong baseline algorithms on real-world datasets.
A new algorithm for deep Q-learning with robustness to state transition uncertainty.
problem Model uncertainty in state transitions for non-tabular, continuous state spaces.
method Distributionally robust approach using worst-case transition ball and dualized Bellman operator with Sinkhorn distance.
result Optimal policy found through solving non-linear Bellman equation with neural network parameterization.
For mass spectra acquired from cancer patients by MALDI or SELDI techniques, automated discrimination between cancer types or stages has often been implemented by machine learnings. These techniques typically generate "black-box" classifiers, which are difficult to interpret biologically. We develop new and efficient s…
Multi-task learning is a learning paradigm which seeks to improve the generalization performance of a learning task with the help of some other related tasks. In this paper, we propose a regularization formulation for learning the relationships between tasks in multi-task learning. This formulation can be viewed as a n…
Unified theory for neural scaling laws in hierarchically compositional data.
problem Understanding neural scaling laws in hierarchically compositional data.
method Probabilistic context-free grammars and power-law distributed production rules.
result Unified learning curve behavior for classification and next-token prediction tasks.
Meta RL learns task structure from experience.
problem Designing efficient reinforcement learning algorithms.
method Separately learns policy and task belief using privileged information.
result Effective at solving complex meta-RL environments.
New method uses multi-task learning to improve molecule representations.
problem Cost, bias, and data requirements in chemical representation generation.
method Intelligent task selection in deep multitask networks with transfer learning.
result Deep representations capture more expressive task-based information.
GraphPPD models graph-level uncertainty using GNN embeddings.
problem Capturing uncertainty in graph-level predictions.
method Variational modelling for posterior predictive distribution.
result Effective uncertainty-aware predictions on graph-level tasks.
We propose an adaptive optimization method for deep learning that dynamically adjusts batch size.
problem Optimizing deep learning models with varying sensitivity to batch size selection.
method Adaptive regularization with dynamically determined stochastic batch size based on gradient norms.
result Our method outperforms state-of-the-art optimization algorithms in generalization and robustness.
FixyNN improves energy efficiency of mobile computer vision tasks.
problem High energy consumption of state-of-the-art CNN models on mobile devices.
method Fixed-weight feature extractor and programmable CNN accelerator for transfer learning.
result Achieved up to 26.6 TOPS/W energy efficiency, nearly 2x more efficient than conventional accelerators.
Deep learning advances impact computer architecture and chip design.
problem Improving system accuracy across various tasks.
method Advances in artificial neural networks and machine learning.
result Future models will be sparsely activated and dynamically routed.
The paper explores how disentangled representations can improve fairness in prediction tasks.
problem Improving fairness in prediction tasks using disentangled representations.
method Investigates different notions of disentanglement and analyzes representations of state-of-the-art models.
result Disentanglement scores are correlated with increased fairness in prediction tasks.
MoleculeSTM learns from molecule structures and texts for better drug design.
problem Lack of integration between chemical structures and textual knowledge in AI drug discovery.
method Jointly learns chemical structures and texts via contrastive learning, using a large dataset.
result MoleculeSTM achieves state-of-the-art performance in zero-shot tasks like structure-text retrieval and molecule editing.
Parallelizes graph embedding for large graphs.
problem Large graphs make existing graph embedding techniques inefficient.
method Distributed parallel computation framework using a cluster of compute nodes.
result Parallel computation scales well and maintains embedding quality.
Transformers struggle to learn Markovian dynamics, showing NP-hard optimization challenges.
problem Understanding transformers' limitations in learning Markovian dynamical functions.
method Investigated through a structured ICL setup, analyzing loss landscapes and parameter optimization.
result Recovering optimal transformer parameters for Markovian functions is NP-hard.
New RBM method for missing data inference, comparing performance to existing methods.
problem Missing data inference and perception-distortion trade-off.
method Linearization of RBM effective energy function for missing data.
result Proposed method outperforms existing reconstruction procedures in missing data inference.
IDS integrates physics engines into deep learning for efficient, interpretable system identification.
problem Lack of generalization and interpretability in learning-based models of physical systems.
method Interactive Differentiable Simulation (IDS) that allows efficient, accurate inference of physical properties.
result Automatic task-based robot design and parameter estimation for nonlinear dynamical systems.
New method for continual learning without task boundaries.
problem Traditional continual learning is task-based and impractical for real-world applications.
method Developed an online continual learning system using Memory Aware Synapses.
result Valid approach demonstrated in self-supervised learning and robot collision avoidance.
Model learns tool affordances from vision, enabling tool selection.
problem Learning tool affordances from visual input.
method Vision-based generative model with task predictor.
result Agents can select appropriate tools based on task success criteria.
Meta-learning improves with explicit modeling of task covariate distributions.
problem Ignoring the relationship between task covariates and conditional distributions limits meta-learning performance.
method Introducing a hierarchical Bayesian model that leverages samples from the marginal task covariates to better infer optimal parameters.
result Our method outperforms initialization-based meta-learning on popular classification benchmarks.
A new meta-learning framework that assigns weights to source tasks based on target samples.
problem Learning initialization for target tasks with limited labeled examples.
method A general framework that assigns weights to the loss of different source tasks, which can depend on the target samples. Provides upper bounds and develops a learning algorithm based on minimizing the error bound with respect to an empirical IPM.
result Empirically, the weighted meta-learning algorithm finds better initializations than uniformly-weighted meta-learning algorithms.
This work improves deep learning models for fMRI by generating realistic brain morphology images.
problem Limited dataset sizes for functional MRI limit the accuracy of deep learning models.
method Proposes a method to generate new fMRI images with realistic brain morphology.
result Demonstrates a 26% improvement in predicting antidepressant treatment response using augmented images.
Abstractor enhances Transformers for relational reasoning, improving sample efficiency and performance.
problem Improving sample efficiency and performance in relational tasks.
method Introduces Abstractor module with relational cross-attention to enable explicit relational reasoning.
result Dramatic improvements in sample efficiency and performance on various relational tasks.