Bayesian deep learning integrates perception and inference for higher-level intelligence.
problem Integrating perception and inference for tasks requiring higher-level intelligence.
method Unified probabilistic framework combining deep learning and Bayesian models.
result Integrating perception and inference leads to improved performance in tasks like recommender systems and topic models.
This research explores inductive biases for deep learning to improve AI's higher-level cognition.
problem Current AI struggles with flexible out-of-distribution and systematic generalization.
method Examines and proposes new inductive biases for deep learning.
result Identifies specific inductive biases for higher-level sequential processing.
Bayesian deep learning integrates deep learning and probabilistic models for better inference.
problem Higher-level inference using probabilistic models is more powerful than deep learning alone.
method Unified probabilistic framework combining deep learning and Bayesian models.
result Bayesian deep learning enhances both perception and inference processes.
This thesis tackles continual learning in neural networks, overcoming forgetting.
problem Overcoming catastrophic forgetting in neural networks.
method Task incremental and online continual learning settings; proposed methods for mitigating forgetting.
result Advances in continual learning state of the art and challenges for application.
Reviews Gaussian Markov models for conditional independence.
problem Understanding conditional independence in probabilistic models.
method Historical review of model selection and estimation techniques.
result Gaussian Markov models are similar but not equivalent to each other.
HiPPO adapts skills and higher-level policies together for better transfer in hierarchical RL.
problem Sub-optimality in skill transfer when lower-level skills are fixed.
method HiPPO: a novel hierarchical policy gradient method that trains all levels of the hierarchy jointly.
result Improved robustness of skills to environment changes through training time-abstractions.
Enhances reinforcement learning with hierarchical policies using latent variables.
problem Improving performance in reinforcement learning tasks with hierarchical policies.
method Training each layer of a hierarchical neural network to solve tasks directly, with latent variables controlling lower layers.
result Improves performance on standard benchmark tasks and complex sparse-reward tasks.
A network supporting deep unsupervised learning is presented. The network is an autoencoder with lateral shortcut connections from the encoder to decoder at each level of the hierarchy. The lateral shortcut connections allow the higher levels of the hierarchy to focus on abstract invariant features. While standard auto…
LYRICS integrates deep learning and logic inference for complex decision-making.
problem Combining deep learning with symbolic logic for intelligent decision-making.
method LYRICS provides a generic interface layer that integrates TF models with FOL knowledge, converting constraints into optimization problems.
result LYRICS enables learning under logical constraints, improving model performance and decision-making.
New framework identifies causal models with arbitrary interventions, improving realism.
problem Identify causal models with realistic interventions.
method Theoretical framework for identifying causal models with arbitrary interventions.
result Identify causal models with arbitrary interventions, up to a higher-level abstraction.
AuGMEnT network struggles with long-term memory for hierarchical tasks.
problem Learning and memory in neural networks, especially hierarchical tasks.
method Introduced hybrid AuGMEnT with leaky and non-leaky memory units.
result Hybrid AuGMEnT solves hierarchical and distractor tasks.
Generative model for scalable vector graphics captures font design statistics.
problem Lack of higher-level understanding in vision and imagery modeling.
method Sequential generative models of vector graphics.
result Model captures statistical dependencies and richness of font datasets.
Factor analysis provides linear factors that describe relationships between individual variables of a data set. We extend this classical formulation into linear factors that describe relationships between groups of variables, where each group represents either a set of related variables or a data set. The model also na…
A new method for solving complex sequential decision-making problems by decomposing them into multiple levels.
problem Sequential decision-making with natural multi-level structure.
method Multi-level meta-reinforcement learning with skill-based curriculum.
result Efficiently reduces stochasticity and policy search space, leading to fewer iterations and computations.
AI learns market manipulation through simulation, suggesting regulation.
problem Regulating AI to prevent market manipulation.
method Used a genetic algorithm in an artificial market simulation.
result AI discovered market manipulation as an optimal strategy.
This work improves generative models by using feedback from multiple dependent models.
problem Improving the performance of generative models in multi-agent systems.
method Building a hierarchical set-up of multiple dependent generative models and using feedback to improve lower-level models.
result The technique improves the performance of lower-level generative models under certain conditions.
Mathematical framework using Riemannian geometry for intelligence and consciousness.
problem Lack of a unified mathematical framework for intelligence and consciousness.
method Conceptualizes intelligence as tokens in a high-dimensional space, using Riemannian geometry to describe structure and dynamics.
result Integrates geometric concepts to offer a unified framework for intelligence and consciousness.
Improved deep learning models using new attribution priors and expected gradients.
problem Improving interpretability and performance of deep learning models.
method Introducing new attribution priors and expected gradients method that satisfies interpretability axioms.
result Improves model performance across various real-world tasks.
Artificial intelligence has impacted many aspects of human life. This paper studies the impact of artificial intelligence on economic theory. In particular we study the impact of artificial intelligence on the theory of bounded rationality, efficient market hypothesis and prospect theory.
Introduces Motion Programs for better video analysis of human motion.
problem Current video analysis focuses on raw pixels or keypoints, missing higher-level motion primitives.
method Introduces Motion Programs as a neuro-symbolic representation of motions as a composition of high-level primitives.
result Motion Programs accurately describe diverse human motions and improve downstream tasks.
The paper proposes a test for trading intelligence similar to the Turing Test.
problem Determining if trading programs exhibit intelligent behavior.
method Develops a test based on the Turing Test for financial trading programs.
result Introduces a method to assess trading intelligence in financial markets.
Proposes methods to improve hierarchical classification accuracy by flattening inconsistent nodes.
problem Error propagation in top-down hierarchical classification due to inconsistent nodes.
method Data-driven approaches for identifying and flattening inconsistent nodes.
result Improves classification performance by up to 7% in Macro-F1 score.
This thesis explores emergent intelligence in disordered systems like spin glasses and neural networks.
problem Understanding the principles behind emergent intelligent behaviors in disordered systems.
method Statistical physics approach to charting learning mechanisms and dynamics.
result Uncovering relationships between learning mechanisms and physical dynamics.
Framework detects cyber threats from Twitter tweets.
problem Time-consuming manual extraction of cyber threat intelligence.
method Novelty detection model trained on CVE data.
result F1-score of 0.643 for classifying cyber threat tweets.
Explains AI basics and its neural origins.
problem Understanding AI and neural origins.
method Overview of AI and its biological roots.
result Highlights the evolution of AI models.
Survey on AI math foundations, focusing on neural networks.
problem Lack of rigorous mathematical foundation for AI.
method Survey and discussion of theoretical directions in AI.
result Discussion of open problems in AI math.
Research designs an AI system to classify malware under adversarial conditions.
problem Adversarial attacks on malware classification algorithms.
method Machine learning-based intelligent systems approach.
result Robust malware classification model under adversarial conditions.
Improved spoken English intelligibility with computer recognition and feature extraction.
problem Improving spoken English pronunciation and intelligibility.
method Automatic speech recognition using PocketSphinx alignment and feature extraction with SVM classifier probability prediction.
result SVM models achieve 82 percent agreement with human transcriptions, up from 75 percent.
Neural network learns to focus on slowly varying latent parameters in time series data.
problem Extracting slowly varying latent parameters from time series data.
method Transform raw data into a higher-level categorical representation, train predictor from new time series to future, introduce non-triviality measure.
result Creates a coarse-grained model focusing on latent parameters, capable of semi-supervised learning.
This study compares feature extraction methods using Neural Networks and Latent Dirichlet Allocation for movie synopses.
problem Extracting meaningful features from movie synopses for pattern detection and recommendation.
method Employed Latent Dirichlet Allocation for topic modeling and Neural Networks for distributed paragraph representations.
result Latent Dirichlet Allocation can provide meaningful features for movie synopses, comparable to Neural Networks.
AI makes better decisions, leading to more rational markets.
problem How AI affects decision-making in markets.
method Investigates how AI makes decisions in markets compared to humans.
result AI decisions are more consistent and rational, making markets more rational.
Survey on deep neural networks for vision and speech.
problem Improving intelligent vision and speech systems.
method Review of deep learning models and challenges.
result Emerging technologies show promise for future systems.
StackNet predicts fluid intelligence from brain images of adolescents.
problem Predicting fluid intelligence in adolescents using brain imaging.
method Feature extraction, normalization, denoising, selection, StackNet architecture, 11 models, 3 layers, 10-fold cross-validation.
result StackNet achieves mean squared errors of 82.42 on training/validation and 94.25 on testing.
Intelligence emerges from stabilizing invariant cycles in memory.
problem Understanding the nature of intelligence and its emergence.
method Structural-dynamical account rooted in a topological closure law: \(\partial^2=0\).
result Memory-amortized inference (MAI) mechanism that implements SbS \(
ightarrow\) CCUP.
Intelligent synapses help neural networks learn continuously.
problem Continual learning in changing data environments.
method Intelligent synapses that accumulate and reuse task-relevant information.
result Significant reduction in forgetting with improved computational efficiency.
Stochasticity is key for machine learning's robustness and generalizability.
problem Machine learning's need for robustness and generalizability.
method Review of ML literature and biological intelligence.
result Stochasticity is a critical ingredient for intelligent systems in ML.
This study compares transfer learning and multi-agent learning for AI-driven traffic agents.
problem Improving traffic flow in mixed-intelligence highway scenarios.
method Online MIT DeepTraffic simulation, deep reinforcement learning, elitist evolutionary algorithm, hyperparameter search, transfer learning, multi-agent learning.
result Transfer learning and multi-agent learning yield different average speeds for AI-driven traffic agents.
This paper tackles URLLC in 6G networks with deep learning.
problem Stringent requirements on end-to-end delay and reliability for mission-critical applications.
method Develops a multi-level architecture combining theoretical models and real-world data, using deep transfer learning and federated learning.
result Demonstrates improved performance in URLLC for mission-critical applications.
FCN improves lidar cloud detection accuracy.
problem Segmenting lidar imagery into cloud locations.
method Semi-supervised learning with pre-training and fully supervised learning.
result FCN achieves higher cloud identification accuracy.
Chess engines Stockfish and LCZero differ in their approach to solving endgame puzzles.
problem Comparing machine and human chess problem-solving abilities.
method Used Plaskett's Puzzle to compare Stockfish and LCZero's performance.
result Stockfish outperforms LCZero on the puzzle.
We are working to develop automated intelligent agents, which can act and react as learning machines with minimal human intervention. To accomplish this, an intelligent agent is viewed as a question-asking machine, which is designed by coupling the processes of inference and inquiry to form a model-based learning unit.…
Future autonomous systems need reliable world models and complex action sequences.
problem Current automated systems lack reliable world models and complex action sequences.
method Introduce energy-based and latent variable models combined in a hierarchical joint embedding predictive architecture (H-JEPA).
result Combining energy-based and latent variable models in H-JEPA can lead to reliable world models and complex action sequences.
Symmetry principles help in creating better AI representations.
problem Creating efficient and generalizable AI representations.
method Using symmetry transformations to guide representation learning.
result Symmetry principles improve data efficiency and generalizability in AI.
This paper discusses issues in mining user behavioral rules for context-aware mobile apps.
problem Mining contextual behavioral rules from smartphone data.
method Addressing quality of data, relevancy of contexts, discretization, rule discovery, semantic understanding, and dynamic rule updating.
result Potential solutions for mining user behavioral rules for context-aware mobile apps.
The paper proposes an AI and IIoT framework for improved maintenance.
problem Current maintenance practices need improvement with AI and IIoT.
method Review of reliability modeling, introduction of Intelligent Maintenance framework, and novel probabilistic deep learning approach.
result Demonstrated novel probabilistic deep learning reliability modelling in Turbofan Engine Degradation Dataset.
Paper tackles missing data in high-dimensional datasets using deep learning and swarm intelligence.
problem Missing data in high-dimensional datasets, considering different mechanisms.
method Combines deep learning and swarm intelligence to estimate missing data.
result Promising approach to estimate missing data, despite longer running times.
The paper shows how uncertainty quantification improves counterfactual explainability in AI.
problem Lack of foundational concepts in transparency research.
method Integrates uncertainty quantification into counterfactual explainability.
result Demonstrates competitive performance of an uncertainty-based explainer.
Computational Intelligence (CI) is a sub-branch of Artificial Intelligence paradigm focusing on the study of adaptive mechanisms to enable or facilitate intelligent behavior in complex and changing environments. There are several paradigms of CI [like artificial neural networks, evolutionary computations, swarm intelli…