Robots learn quickly from few interactions using mental replay and intrinsic motivation.
problem Continuous online adaptation for robots in changing environments.
method Bio-inspired stochastic recurrent neural network with learning signals and mental replay.
result Robots can adapt to novel environments in seconds from few interactions.
A new method backtracks through a few key past states to speed up credit assignment in long sequences.
problem Computational inefficiency of back-propagation through time for long sequences.
method Sparse attentive backtracking using learned attention mechanisms to skip connections.
result Matches or outperforms regular BPTT and truncated BPTT in tasks with long-term dependencies.
Deep learning predicts mental disorders from audio and text samples.
problem Predicting mental disorders from speech samples.
method Multimodal deep learning structure using various pre-trained models for audio and text embeddings, transfer learning, and auxiliary corpora.
result Acceptable accuracy in predicting mental disorders through multimodal analysis.
This research improves interpretability in sequential explanations using mental models.
problem Improving interpretability in sequential explanations between two parties.
method A reinforcement learning framework that selects explanations based on the explainee's mental model.
result Mental model-based policies increase interpretability over random selection in multiple sequential explanations.
A novel framework optimizes experience replay for reinforcement learning.
problem Optimizing the replay policy for better performance in off-policy reinforcement learning.
method Proposes a novel ERO framework that alternately updates the agent policy and replay policy.
result Demonstrates improved performance in various continuous control tasks.
Framework integrates mental disorder measurements for personalized treatment.
problem Optimizing treatment for mental disorders with latent mental states and heterogeneity.
method Measurement theory and multi-layer neural network for complex treatment effects.
result Learned treatment policies outperform alternatives on heterogeneous treatment effects.
The study predicts how discussions in mental disorder Reddit communities affect users' emotional states.
problem Improving mental health conditions through social support analysis.
method Text embedding techniques and RNNs for predicting emotional tone shifts.
result Users' emotional states can improve due to social support, as evidenced by positive comments following negative posts.
Reverse Experience Replay improves Deep Q-learning for sparse rewards.
problem Sparse rewards and reward-maximizing tasks in Deep Q-learning.
method Sampling transitions in reverse order for training.
result Significantly increased performance in tasks with limited experience and memory capacity.
Combines experience replay techniques to improve reinforcement learning.
problem Improving reinforcement learning algorithms.
method Combines CER, PER, and HER with DDPG and DQN.
result Effective combinations of these techniques in various environments.
New modifiers improve noisy RNN replay in hippocampal networks.
problem Improving noisy RNN replay in hippocampal networks.
method Three approaches: hidden state leakage, adaptation, and momentum.
result Hidden state leakage, adaptation, and momentum improve noisy RNN replay.
New insights into experience replay in RL algorithms.
problem Understanding the impact of replay capacity and replay ratio in Q-learning.
method Systematic and extensive analysis of experience replay in Q-learning methods, focusing on replay capacity and replay ratio.
result Greater replay capacity significantly improves performance for certain algorithms, while other techniques offer limited benefit.
More than two thirds of mental health problems have their onset during childhood or adolescence. Identifying children at risk for mental illness later in life and predicting the type of illness is not easy. We set out to develop a platform to define subtypes of childhood social-emotional development using longitudinal,…
New method improves continual learning by generating samples and labels together.
problem Improving continual learning in complex classification tasks.
method Conditional replay generates samples and labels together, reducing label inference error.
result Conditional replay outperforms marginal replay and EWC on benchmarks.
Semiparametric STAR model improves mental health data analysis.
problem Overdispersed, zero-inflated, bounded count data in self-reported mental health surveys.
method STAR transformation and rounding of latent Gaussian model, nonparametric transformation estimation, EM algorithm for maximum likelihood.
result Substantial improvements in goodness-of-fit compared to existing models.
Research uses activity analysis to identify mental health symptoms.
problem Identifying mental health symptoms using objective activity metrics.
method Proposes a framework for mHealth monitoring of psychiatric patients based on physical activity time series.
result Identifies distinct behavioural phenotypes and measures for mood assessment.
Dynamic topic model improves mental health note analysis for children.
problem Lack of longitudinal topic models for psychiatric clinical notes.
method Developed a dynamic topic model with consistent topics and individualized temporal dependencies.
result Achieved a 38% increase in topic coherence.
Curious Replay improves model-based reinforcement learning agents' adaptability.
problem Existing model-based reinforcement learning agents struggle to adapt quickly to changing environments.
method Curious Replay uses a curiosity-based priority signal for prioritized experience replay tailored to model-based agents.
result Agents using Curious Replay achieve improved performance in exploration and on benchmarks.
New dataset tests mental rotation from single images, improving model understanding of 3D scenes.
problem Understanding how a scene looks from a different viewpoint using a single image.
method Created CLEVR-MRT dataset, explored neural architectures for volumetric scene representations.
result Demonstrated the effectiveness of volumetric representations in answering mental rotation questions.
Estimates statistical shifts across subjects for EEG-based mental workload assessment.
problem Variability in EEG correlates of mental workload across subjects makes model generalization difficult.
method Proposes a strategy to estimate marginal and conditional shifts between multiple data distributions.
result Estimates of statistical shifts can improve mental workload prediction accuracy.
Improved speaker verification system detects replay attacks.
problem Detecting replay attacks in speaker verification systems.
method Multi-task learning of noise classes (replay noise, recording environment, recording device noises) and spoofing detection.
result Performance improved by 30% on ASVspoof2017 datasets.
A new method prioritizes and recycles experiences for better reinforcement learning.
problem Improving reinforcement learning efficiency by prioritizing and recycling experiences.
method Double-prioritized state-recycled (DPSR) experience replay.
result DPSR achieved state-of-the-art results in Atari games, outperforming original and prioritized methods.
ReaPER improves learning efficiency by prioritizing reliable experiences.
problem Inefficient sampling of past experiences in reinforcement learning.
method Introducing a novel measure of reliability to prioritize experiences in PER.
result ReaPER outperforms PER in various environments, including Atari-10.
Unified study of stateful replay for streaming learning, reducing forgetting by 2-3x.
problem Catastrophic forgetting in streaming generative and predictive learning.
method Unified analysis of stateful replay for autoencoding, forecasting, and classification tasks.
result Stateful replay reduces average forgetting by a factor of 2-3 on heterogeneous multi-task streams.
Study integrates diverse data sources to predict mental health conditions.
problem Predict individuals' mental health conditions using a heterogeneous network approach.
method Leverage a heterogeneous information network (HIN) to model social interaction, health data, and survey data. Apply recommender system (RS) and node classification (NC) paradigms to predict mental health states.
result RS and NC methods outperform traditional logistic regression models in predicting mental health conditions.
Machine learning detects and classifies cognitive distortions in mental health texts.
problem Detect and classify cognitive distortions in mental health texts to improve treatment.
method Machine learning framework using crowdsourced and therapy dataset.
result Model achieved high F1 scores for detecting and classifying cognitive distortions.
A framework for goal-based investing with penalties for fund transfers.
problem Investors' mental accounting and multiple investment goals.
method Continuous-time portfolio selection with mental costs and penalties.
result The value function is the unique solution to a complex system of equations.
Combines experience replay and exploration for better agent performance.
problem Improving exploration efficiency and robustness in reinforcement learning.
method Integrates Intrinsic Rewards with Prioritized Oversampled Experience Replay (POER).
result Achieves better agent performance and sample efficiency compared to PPO/RND.
Generative replay improves continual learning by using generated data as negative examples.
problem Catastrophic forgetting in continual learning.
method Using generative models to provide negative examples for new classes.
result Generative replay can improve learning new classes even when existing approaches fail.
Study identifies mental stress in firefighters using heart rate variability data.
problem Unsupervised identification of mental stress in firefighters from heart rate variability data.
method Exploration and comparison of three unsupervised methods: K-Means, convolutional autoencoders, and LSTM autoencoders.
result Convolutional and LSTM autoencoders successfully stratify stressed versus normal samples using HRV markers.
Model predicts depression and anxiety from social network dynamics.
problem Predicting mental health from social interactions.
method Developed a predictive model using dynamic social network data.
result Dynamic social network model outperforms static models.
Automated EEG analysis gauges mental workload in task evaluation.
problem Evaluating mental workload in user tasks, especially complex ones.
method Experimental setup, EEG data analysis, feature extraction, machine learning.
result Machine learning outperforms traditional methods in mental workload estimation.
Efficient actor-critic learning with shared experience replay improves data efficiency.
problem Challenges in actor-critic reinforcement learning with experience replay and off-policy learning stability.
method Combining actor-critic algorithms with shared experience replay, analyzing V-trace, proposing a trust region scheme.
result State-of-the-art data efficiency on Atari achieved with 200M environment frames.
This work adapts RDT for mental program construction, showing benefits and costs.
problem Applying RDT to mental programs with trade-offs between description length, error, and computational costs.
method Proposed a three-way trade-off and used simulations and partial information decomposition.
result Constructing a shared program library provides global benefits but is sensitive to curricula.
The paper improves reinforcement learning stability and efficiency with a new theoretical framework.
problem Stability and efficiency in reinforcement learning, especially in data-scarce scenarios.
method Theoretical framework using resampled U- and V-statistics to model experience replay, applied to policy evaluation and kernel ridge regression. result Significant improvements in stability and efficiency, particularly in data-scarce scenarios.
The paper studies how memory replay affects reinforcement learning performance.
problem Understanding the effects of memory size and prioritization in experience replay.
method Formulated a dynamical systems ODE model of Q-learning with experience replay, derived analytic solutions, and proposed an adaptive memory buffer size algorithm.
result The amount of memory kept and prioritization significantly impact learning dynamics; too much or too little memory slows down learning.
A new method for prioritizing sequences in reinforcement learning improves learning efficiency.
problem Efficient learning in reinforcement learning algorithms.
method Prioritized Sequence Experience Replay (PSER) for prioritizing sequences of experience.
result PSER is guaranteed to converge faster than PER and substantially improves performance.
A new system detects audio replay attacks with high accuracy.
problem Detecting and preventing audio replay attacks in speaker verification systems.
method Proposes Attentive Filtering Network combining attention-based filtering and ResNet classifier.
result Achieves EER of 8.99% on ASVspoof 2017 Version 2.0 dataset.
A new method improves continual learning by replaying pseudo data and using orthogonal weight modification.
problem Catastrophic forgetting in class incremental learning.
method Generative replay with orthogonal weight modification.
result Our method achieves substantial improvement over conventional generative replay and OWM.
Algorithm creates synthetic experiences to enhance Deep Reinforcement Learning.
problem Lack of synthetic experiences in classic Experience Replay.
method Bootstraps synthetic experiences to augment the replay memory.
result Synthetic experiences can improve learning speed and performance.
A new memory replay mechanism improves reinforcement learning stability and speed.
problem Forgetting in reinforcement learning with continuous control.
method Augmented Memory Replay (AMR) that optimizes the replay of past experiences.
result AMR enhances stability and convergence speed of learning algorithms.
New method combines brain imaging data from multiple studies to improve cognitive decoding.
problem Low statistical power in individual neuroimaging studies.
method A new methodology to analyze brain responses across tasks without a unified theoretical framework.
result Improves decoding performance for 80% of 35 functional-imaging studies.
Generative replay with feedback connections improves continual learning efficiency.
problem Catastrophic forgetting in artificial neural networks when learning new tasks.
method Split and permuted MNIST tasks, generative replay with distillation, and feedback connections.
result Generative replay with distillation outperformed other methods in all scenarios.
ReF-ER improves RL by selectively updating experiences based on policy similarity.
problem Deterioration of experience replay accuracy when policies diverge.
method Enforces policy similarity in replay memory by skipping unlikely experiences and regulating policy changes.
result ReF-ER consistently improves RL performance across various benchmarks.
Prototype-based generative replay framework for online continual regression.
problem Addressing the challenge of non-stationary data streams in regression tasks.
method Adaptive output-space discretization model for prototype-based generative replay.
result Reduces forgetting and provides more stable performance.
New framework models echo chamber learning, proving tight bounds on algorithm performance.
problem Echo chambers in machine learning where systems learn from self-annotated data.
method Online Learning in the Replay Setting, Extended Threshold dimension, closure-based learner.
result Proves tight bounds on algorithm performance against replay adversaries.
New method retrieves most interfered samples for continual learning.
problem Challenges in continual learning with online data streams.
method Controlled sampling of most interfered samples for replay.
result Consistent gains in performance and reduced forgetting.
AI identifies patient clusters for diabetes case management.
problem Diabetes complications and mental health comorbidities drive high healthcare costs.
method Combined AI techniques with diverse data sources for prediction and clustering.
result 83.5% accuracy in predicting diabetes complications and meaningful patient clusters.
Study shows high-rise buildings in Dhaka affect mental health, especially lower-income residents.
problem Mental health risks due to high-rise buildings in Dhaka.
method Computer vision pipeline to analyze sky visibility, greenery, and colors in streets.
result Lower-income residents suffer more from lack of sky visibility and greenery in their environment.