Search Results for author:
Found 58 papers, 27 papers with code
Understanding R1-Zero-Like Training: A Critical Perspective
DeepSeek-R1-Zero has shown that reinforcement learning (RL) at scale can directly enhance the reasoning capabilities of LLMs without supervised fine-tuning.
EVaDE : Event-Based Variational Thompson Sampling for Model-Based Reinforcement Learning
In this paper, we propose Event-based Variational Distributions for Exploration (EVaDE), which are variational distributions that are useful for MBRL, especially when the underlying domain is object-based.
Model-based Reinforcement Learning
reinforcement-learning
+2
Sample-Efficient Alignment for LLMs
The results demonstrate that SEA achieves highly sample-efficient alignment with oracle's preferences, outperforming recent active exploration methods for LLMs.
Hierarchical Neural Constructive Solver for Real-world TSP Scenarios
Existing neural constructive solvers for routing problems have predominantly employed transformer architectures, conceptualizing the route construction as a set-to-sequence learning task.
Differentiable Cluster Graph Neural Network
Graph Neural Networks often struggle with long-range information propagation and in the presence of heterophilous neighborhoods.
Lightweight Spatial Modeling for Combinatorial Information Extraction From Documents
We propose KNN-former, which incorporates a new kind of spatial bias in attention calculation based on the K-nearest-neighbor (KNN) graph of document entities.
On the Empirical Complexity of Reasoning and Planning in LLMs
Chain-of-thought (CoT), tree-of-thought (ToT), and related techniques work surprisingly well in practice for some complex reasoning tasks with Large Language Models (LLMs), but why?
Continual Learning of Numerous Tasks from Long-tail Distributions
In this paper, we investigate the performance of continual learning algorithms with a large number of tasks drawn from a task distribution that is long-tail in terms of task sizes.
Constrained Layout Generation with Factor Graphs
This paper addresses the challenge of object-centric layout generation under spatial constraints, seen in multiple domains including floorplan design process.
PF-GNN: Differentiable particle filtering based approximation of universal graph representations
Message passing Graph Neural Networks (GNNs) are known to be limited in expressive power by the 1-WL color-refinement test for graph isomorphism.
Locality Sensitive Sparse Encoding for Learning World Models Online
Unfortunately, NN-based models need re-training on all accumulated data at every interaction step to achieve FTL, which is computationally expensive for lifelong agents.
Differentiable Tree Search Network
In this work, we introduce Differentiable Tree Search Network (D-TSN), a novel neural network architecture that significantly strengthens the inductive bias by embedding the algorithmic structure of a best-first online search algorithm.
Tell2Design: A Dataset for Language-Guided Floor Plan Generation
We make multiple contributions to initiate research on this task.
Factor Graph Neural Networks
In recent years, we have witnessed a surge of Graph Neural Networks (GNNs), most of which can learn powerful representations in an end-to-end fashion with great success in many real-world applications.
Efficient Offline Policy Optimization with a Learned Model
Instead of planning with the expensive MCTS, we use the learned model to construct an advantage estimation based on a one-step rollout.
Differentiable Parsing and Visual Grounding of Natural Language Instructions for Object Placement
Natural language generally describes objects and spatial relations with compositionality and ambiguity, two major obstacles to effective language grounding.
None Class Ranking Loss for Document-Level Relation Extraction
This ignores the context of entity pairs and the label correlations between the none class and pre-defined classes, leading to sub-optimal predictions.
Document-level Relation Extraction
Emotion Classification
+4
Graph Representation Learning with Individualization and Refinement
Individualization refers to artificially distinguishing a node in the graph and refinement is the propagation of this information to other nodes through message passing.
Combining Reinforcement Learning and Optimal Transport for the Traveling Salesman Problem
This paper exemplifies the integration of entropic regularized optimal transport techniques as a layer in a deep reinforcement learning network.
Context-Hierarchy Inverse Reinforcement Learning
To solve these issues, we present Context Hierarchy IRL(CHIRL), a new IRL algorithm that exploits the context to scale up IRL and learn reward functions of complex behaviors.
ExPoSe: Combining State-Based Exploration with Gradient-Based Online Search
Depending upon the smoothness of the action-value function, one approach to overcoming this issue is through online learning, where information is interpolated among similar states; Policy Gradient Search provides a practical algorithm to achieve this.
Ensemble and Auxiliary Tasks for Data-Efficient Deep Reinforcement Learning
Ensemble and auxiliary tasks are both well known to improve the performance of machine learning models when data is limited.
Learning Latent Graph Dynamics for Visual Manipulation of Deformable Objects
Manipulating deformable objects, such as ropes and clothing, is a long-standing challenge in robotics, because of their large degrees of freedom, complex non-linear dynamics, and self-occlusion in visual perception.
State-Aware Variational Thompson Sampling for Deep Q-Networks
We derive a variational Thompson sampling approximation for DQNs which uses a deep network whose parameters are perturbed by a learned variational noise distribution.
Factor Graph Molecule Network for Structure Elucidation
Designing a network to learn a molecule structure given its physical/chemical properties is a hard problem, but is useful for drug discovery tasks.
Neuralizing Efficient Higher-order Belief Propagation
In this paper, we propose to combine these approaches to learn better node and graph representations.
Contrastive Variational Reinforcement Learning for Complex Observations
This paper presents Contrastive Variational Reinforcement Learning (CVRL), a model-based method that tackles complex visual observations in DRL.
Understanding and Resolving Performance Degradation in Graph Convolutional Networks
In this work, we study performance degradation of GCNs by experimentally examining how stacking only TRANs or PROPs works.
Multiplicative Gaussian Particle Filter
We propose a new sampling-based approach for approximate inference in filtering problems.
Discriminative Particle Filter Reinforcement Learning for Complex Partial Observations
The particle filter maintains a belief using learned discriminative update, which is trained end-to-end for decision making.
Visual Relationship Detection with Low Rank Non-Negative Tensor Decomposition
We address the problem of Visual Relationship Detection (VRD) which aims to describe the relationships between pairs of objects in the form of triplets of (subject, predicate, object).
An Interactive Multi-Task Learning Network for End-to-End Aspect-Based Sentiment Analysis
Aspect-based sentiment analysis produces a list of aspect terms and their corresponding sentiments for a natural language sentence.
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+4
Factor Graph Neural Network
Most of the successful deep neural network architectures are structured, often consisting of elements like convolutional neural networks and gated recurrent neural networks.
Particle Filter Recurrent Neural Networks
Recurrent neural networks (RNNs) have been extraordinarily successful for prediction with sequential data.
LeTS-Drive: Driving in a Crowd by Learning from Tree Search
LeTS-Drive leverages the robustness of planning and the runtime efficiency of learning to enhance the performance of both.
Differentiable Algorithm Networks for Composable Robot Learning
This paper introduces the Differentiable Algorithm Network (DAN), a composable architecture for robot learning systems.
Factored Contextual Policy Search with Bayesian Optimization
Our key observation is that experience can be directly generalized over target contexts.
Adaptive Semi-supervised Learning for Cross-domain Sentiment Classification
We consider the cross-domain sentiment classification problem, where a sentiment classifier is to be learned from a source domain and to be generalized to a target domain.
Effective Attention Modeling for Aspect-Level Sentiment Classification
First, we propose a method for target representation that better captures the semantic meaning of the opinion target.
Integrating Algorithmic Planning and Deep Learning for Partially Observable Navigation
We propose to take a novel approach to robot system design where each building block of a larger system is represented as a differentiable program, i. e. a deep neural network.
Push-Net: Deep Planar Pushing for Objects with Unknown Physical Properties
This paper introduces Push-Net, a deep recurrent neural network model, which enables a robot to push objects of unknown physical properties for re-positioning and re-orientation, using only visual camera images as input.
Exploiting Document Knowledge for Aspect-level Sentiment Classification
Attention-based long short-term memory (LSTM) networks have proven to be useful in aspect-level sentiment classification.
PORCA: Modeling and Planning for Autonomous Driving among Many Pedestrians
Our planning system combines a POMDP algorithm with the pedestrian motion model and runs in near real time.
Robotics
Particle Filter Networks with Application to Visual Localization
Particle filtering is a powerful approach to sequential state estimation and finds application in many domains, including robot localization, object tracking, etc.
Convolutional Sequence to Sequence Model for Human Dynamics
Human motion modeling is a classic problem in computer vision and graphics.
HyP-DESPOT: A Hybrid Parallel Algorithm for Online Planning under Uncertainty
Planning under uncertainty is critical for robust robot performance in uncertain, dynamic environments, but it incurs high computational cost.
Intention-Net: Integrating Planning and Deep Learning for Goal-Directed Autonomous Navigation
How can a delivery robot navigate reliably to a destination in a new office building, with minimal prior information?
Tensor Belief Propagation
We propose a new approximate inference algorithm for graphical models, tensor belief propagation, based on approximating the messages passed in the junction tree algorithm.
An Unsupervised Neural Attention Model for Aspect Extraction
Unlike topic models which typically assume independently generated words, word embedding models encourage words that appear in similar contexts to be located close to each other in the embedding space.
QMDP-Net: Deep Learning for Planning under Partial Observability
It is a recurrent policy network, but it represents a policy for a parameterized set of tasks by connecting a model with a planning algorithm that solves the model, thus embedding the solution structure of planning in a network learning architecture.
Factored Contextual Policy Search with Bayesian Optimization
Scarce data is a major challenge to scaling robot learning to truly complex tasks, as we need to generalize locally learned policies over different "contexts".
DESPOT: Online POMDP Planning with Regularization
We show that the best policy obtained from a DESPOT is near-optimal, with a regret bound that depends on the representation size of the optimal policy.
Robustness of Bayesian Pool-based Active Learning Against Prior Misspecification
This suggests we should use a Lipschitz utility for AL if robustness is required.
POMDP-lite for Robust Robot Planning under Uncertainty
We show that a POMDP-lite is equivalent to a set of fully observable Markov decision processes indexed by a hidden parameter and is useful for modeling a variety of interesting robotic tasks.
Adaptive Stochastic Optimization: From Sets to Paths
Adaptive stochastic optimization optimizes an objective function adaptively under uncertainty.
Learning with Invariance via Linear Functionals on Reproducing Kernel Hilbert Space
For the representer theorem to hold, the linear functionals are required to be bounded in the RKHS, and we show that this is true for a variety of commonly used RKHS and invariances.
Active Learning for Probabilistic Hypotheses Using the Maximum Gibbs Error Criterion
We introduce a new objective function for pool-based Bayesian active learning with probabilistic hypotheses.
Monte Carlo Bayesian Reinforcement Learning
Bayesian reinforcement learning (BRL) encodes prior knowledge of the world in a model and represents uncertainty in model parameters by maintaining a probability distribution over them.
