EILearn learns incrementally using previous classifier knowledge.
problem Incremental learning with previous data.
method Retains and uses previous classifier knowledge, monitors performance, eliminates poorly performing classifiers.
result Outperforms existing incremental learning approaches.
Bayesian models improve continual learning with natural and Stein gradients.
problem Bayesian models struggle with catastrophic forgetting in sequential learning.
method Use natural gradients and Stein gradients to facilitate Bayesian continual learning.
result Bayesian models retain knowledge from previous tasks more effectively.
Link prediction in large knowledge graphs has received a lot of attention recently because of its importance for inferring missing relations and for completing and improving noisily extracted knowledge graphs. Over the years a number of machine learning researchers have presented various models for predicting the prese…
Induction of common sense knowledge about prototypical sequences of events has recently received much attention. Instead of inducing this knowledge in the form of graphs, as in much of the previous work, in our method, distributed representations of event realizations are computed based on distributed representations o…
This paper studies Bayesian ranking and selection (R&S) problems with correlated prior beliefs and continuous domains, i.e. Bayesian optimization (BO). Knowledge gradient methods [Frazier et al., 2008, 2009] have been widely studied for discrete R&S problems, which sample the one-step Bayes-optimal point. When used ove…
This paper improves Gaussian process predictions by integrating prior knowledge.
problem Gaussian processes lack predictive power when prior information is ignored.
method Derive mean and covariance functions from previous data using weighted sums of basis functions.
result Integrating prior knowledge significantly increases look-ahead time and accuracy.
MetaCL enables neural networks to learn from small data streams without forgetting.
problem Learning from limited data and adapting to new concepts over time.
method MetaCL trains a model to exploit intrinsic data features and dynamically penalize model parameter changes.
result MetaCL achieves state-of-the-art performance on image classification benchmarks.
Enhances knowledge graph completion with mixed geometry tensor factorization.
problem Capturing nuanced distributional properties in knowledge graphs.
method Combines Euclidean and hyperbolic geometries for tensor factorization.
result Improves link prediction accuracy with fewer parameters.
Deep neural networks are known to suffer the catastrophic forgetting problem, where they tend to forget the knowledge from the previous tasks when sequentially learning new tasks. Such failure hinders the application of deep learning based vision system in continual learning settings. In this work, we present a simple …
When training data is sparse, more domain knowledge must be incorporated into the learning algorithm in order to reduce the effective size of the hypothesis space. This paper builds on previous work in which knowledge about qualitative monotonicities was formally represented and incorporated into learning algorithms (e…
Tensor-based embeddings improve knowledge graph fact prediction.
problem Predicting new facts in knowledge graphs.
method Knowledge-Enriched Tensor Factorization
result 5% to 50% relative improvement over state-of-the-art techniques.
New algorithms reduce regret in unconstrained online learning with unknown parameters.
problem Online convex optimization with unknown Lipschitz constant and comparison point.
method Developed algorithms with polynomial bounds that adapt to unknown parameters.
result Achieved improved regret bounds with polynomial dependence on all parameters.
X-DER improves continual learning by revising replay memory and learning unseen classes.
problem Deep networks forget when learning new tasks sequentially.
method Combines rehearsal and Knowledge Distillation, revising replay memory and learning unseen classes.
result X-DER outperforms state-of-the-art on various benchmarks.
TuckER predicts missing facts in knowledge graphs using tensor decomposition.
problem Predicting missing facts in knowledge graphs.
method TuckER uses Tucker decomposition of binary tensor representations.
result TuckER outperforms state-of-the-art models in link prediction.
HSML tailors meta-learning knowledge to task clusters.
problem Handling task uncertainty and heterogeneity in meta-learning.
method Gradient-based HSML with hierarchical task clustering.
result HSML achieves state-of-the-art performance in few-shot learning.
Survey of methods to incorporate external knowledge into stock price prediction.
problem Challenges in predicting stock prices due to market volatility and non-linearity.
method Survey of methods for acquiring and incorporating external knowledge into stock price prediction models.
result Systematic synthesis of previous studies on external knowledge types and their application in stock price prediction.
Transfer learning improves algorithm recommendation performance.
problem Improving algorithm recommendation with limited data.
method Train a neural network on meta-datasets, then transfer knowledge to similar datasets.
result Transfer learning enhances meta-learning for algorithm recommendation.
MEI model improves knowledge graph completion by efficiently modeling interactions between embeddings.
problem Efficiently modeling interactions between knowledge graph embeddings to predict missing links.
method MEI divides embeddings into partitions and uses Tucker and block term formats to model interactions efficiently.
result Achieves state-of-the-art performance on link prediction tasks.
Incorporation of a new knowledge into neural networks with simultaneous preservation of the previous one is known to be a nontrivial problem. This problem becomes even more complex when new knowledge is contained not in new training examples, but inside the parameters (connection weights) of another neural network. Her…
New agent learns from previous search spaces to improve NAS efficiency.
problem NAS requires restarting learning from scratch between different search spaces.
method Transformer-based agent for joint training and knowledge transfer.
result Efficient knowledge transfer between search spaces improves NAS performance.
Survey updates knowledge on homogeneous Einstein spaces.
problem Understanding homogeneous Einstein spaces.
method Building on previous surveys by Wang and Lauret.
result Current state of homogeneous Einstein spaces.
Proposes a method to identify causal relationships using background knowledge.
problem Identifying causal relationships in the presence of background knowledge.
method Learning local structure using all types of causal background knowledge (direct, non-ancestral, ancestral). Criteria for identifying causal relationships based on local structure.
result Effective and efficient method for local structure learning and causal relationship identification.
CLEAS improves neural architecture search for continual learning.
problem Overcoming catastrophic forgetting and adapting to new tasks while controlling model complexity.
method Neural architecture search (NAS) with reinforcement learning to find optimal neural architecture.
result CLEAS achieves higher classification accuracy with simpler neural architectures.
Modulating masks improve lifelong reinforcement learning.
problem Catastrophic forgetting and task interference in lifelong reinforcement learning.
method Adapted modulating masks for deep lifelong reinforcement learning (LRL) with PPO and IMPALA agents.
result Superior performance in both discrete and continuous RL tasks compared to LRL baselines.
This work integrates domain knowledge into A*-based causal discovery methods.
problem Efficiently incorporating domain knowledge into A*-based causal discovery methods.
method Integrates various types of domain knowledge into A*-based causal discovery methods, reducing the graph search space and improving computational gains.
result Small amounts of domain knowledge can dramatically speed up A*-based causal discovery and improve its performance and practicality.
Machine learning improves planetary space physics by incorporating physical knowledge.
problem Improving performance and interpretability of machine learning models for planetary space physics.
method Building on a previous semi-supervised physics-based classification, the team used varying data and physical information to improve machine learning performance and interpretability.
result Incorporating physical knowledge improves machine learning performance and interpretability, essential for deriving scientific meaning.
New PAC-Bayesian framework for flexible meta-learning.
problem Limitations in transferring knowledge between tasks in meta-learning.
method PAC-Bayesian theory applied to learning the learning algorithm.
result Flexibility in meta-learning mechanisms and improved prediction quality.
In many applications of black-box optimization, one can evaluate multiple points simultaneously, e.g. when evaluating the performances of several different neural network architectures in a parallel computing environment. In this paper, we develop a novel batch Bayesian optimization algorithm --- the parallel knowledge…
Knowledge-based planning (KBP) is an automated approach to radiation therapy treatment planning that involves predicting desirable treatment plans before they are then corrected to deliverable ones. We propose a generative adversarial network (GAN) approach for predicting desirable 3D dose distributions that eschews th…
Bayes EMbedding refines embeddings from knowledge and behavior graphs.
problem Combining knowledge and behavior graphs for better entity embeddings.
method Bayesian framework integrating pre-trained embeddings from both graphs.
result BEM refines embeddings from both knowledge and behavior graphs.
A new method for efficient computation of Knowledge Gradient in Bayesian optimization.
problem Efficient computation of the Knowledge Gradient for Bayesian optimization.
method One-shot Hybrid KG, a new approach combining previous ideas.
result The new method is cheap to compute and preserves theoretical properties of previous methods.
Most of previous work in knowledge base (KB) completion has focused on the problem of relation extraction. In this work, we focus on the task of inferring missing entity type instances in a KB, a fundamental task for KB competition yet receives little attention. Due to the novelty of this task, we construct a large-sca…
ARML learns task relations to improve meta-learning efficiency.
problem Handling task heterogeneity in meta-learning.
method Automatically extracts cross-task relations and constructs a meta-knowledge graph.
result ARML outperforms state-of-the-art baselines in few-shot learning tasks.
In this paper, we propose a simple but effective method for training neural networks with a limited amount of training data. Our approach inherits the idea of knowledge distillation that transfers knowledge from a deep or wide reference model to a shallow or narrow target model. The proposed method employs this idea to…
DRUM discovers interpretable rules from knowledge graphs for unseen entities.
problem Inductive link prediction on unseen entities and lack of interpretability.
method Differentiable approach using bidirectional RNNs for low-rank tensor approximation.
result DRUM outperforms existing methods in inductive link prediction.
Enhances neural networks with logical knowledge for better performance.
problem Improving neural network performance with logical knowledge.
method Integrating logical knowledge into neural networks through a new final layer with learnable clause weights.
result KENN outperforms other methods in collective classification tasks with relational data.
A new tensor-based method for predicting temporal relationships in knowledge bases.
problem Predicting temporal relationships in evolving knowledge bases.
method Tensor decomposition of order 4 with new regularization schemes.
result Achieves state-of-the-art performance in temporal link prediction.
Tiny episodic memory significantly improves continual learning performance.
problem Transfer knowledge between tasks in continual learning.
method Store and replay a small number of examples from previous tasks.
result A simple baseline outperforms CL approaches with episodic memory.
Hi-RES framework extracts medical relations from articles and EHRs.
problem Manual annotation bottleneck in relation extraction.
method Labeling sentences, creating improved negative samples, using pretrained language models, and combining EHR embeddings.
result Significant accuracy increases in relation extraction, up to 0.998 for disorder-location relations.
The AI2 Reasoning Challenge (ARC), a new benchmark dataset for question answering (QA) has been recently released. ARC only contains natural science questions authored for human exams, which are hard to answer and require advanced logic reasoning. On the ARC Challenge Set, existing state-of-the-art QA systems fail to s…
Paper tackles robust knowledge transfer in parallel RL tasks.
problem Transfer knowledge from low-tier to high-tier tasks in parallel RL without shared dynamics or reward functions.
method Identifies Optimal Value Dominance condition and proposes online learning algorithms for both tasks.
result Achieves constant regret on partial states and near-optimal regret when tasks are dissimilar.
Convolutional neural networks have achieved astonishing results in different application areas. Various methods that allow us to use these models on mobile and embedded devices have been proposed. Especially binary neural networks are a promising approach for devices with low computational power. However, training accu…
Selective sampling improves matrix completion with known structure.
problem Reconstructing a low-rank matrix with incomplete data.
method Designing observation sets based on matrix structure and selective sampling.
result Improved reconstruction accuracy with selective sampling.
DBULL learns new clusters without forgetting past knowledge in streaming unlabelled data.
problem Challenges in Unsupervised Lifelong Learning with evolving data distributions and class labels.
method Bayesian framework for incremental learning, Deep Bayesian Unsupervised Lifelong Learning (DBULL) algorithm, knowledge preservation mechanism, automatic cluster discovery.
result DBULL can progressively discover new clusters without forgetting past knowledge in unlabelled data.
This paper proposes an online knowledge distillation method that transfers feature map information in addition to class probabilities.
problem Previous online knowledge distillation methods only utilized class probabilities, missing feature map information.
method Adversarial training framework to transfer feature map information; multiple networks trained simultaneously with discriminators.
result Our method performs better than direct alignment methods and is more suitable for online distillation.
Knowledge Matters: Importance of Prior Information for Optimization [7], by Gulcehre et. al., sought to establish the limits of current black-box, deep learning techniques by posing problems which are difficult to learn without engineering knowledge into the model or training procedure. In our work, we completely solve…
In complex transfer learning scenarios new tasks might not be tightly linked to previous tasks. Approaches that transfer information contained only in the final parameters of a source model will therefore struggle. Instead, transfer learning at a higher level of abstraction is needed. We propose Leap, a framework that …
Paper explains how early stopping helps distillation in overparameterized neural networks.
problem Understanding how overparameterized neural networks can improve with early stopping.
method Introducing Anisotropic Information Retrieval (AIR) to justify early stopping in distillation.
result Self-distillation algorithm improves over just early stopping, leading to better generalization.