An official website of the United States government
The paper introduces a new algorithm for planning in partially observable Markov decision processes (POMDP) based on the idea of aggregate simulation. The algorithm uses product distributions to approximate the belief state and shows how to build a representation graph of an approximate action-value function over belief space. The graph captures the result of simulating the model in aggregate under independence assumptions, giving a symbolic representation of the value function. The algorithm supports large observation spaces using sampling networks, a representation of the process of sampling values of observations, which is integrated into the graph representation. Following previous work in MDPs this approach enables action selection in POMDPs through gradient optimization over the graph representation. This approach complements recent algorithms for POMDPs which are based on particle representations of belief states and an explicit search for action selection. Our approach enables scaling to large factored action spaces in addition to large state spaces and observation spaces. An experimental evaluation demonstrates that the algorithm provides excellent performance relative to state of the art in large POMDP problems.
more »
« less
This content will become publicly available on November 7, 2026
State Space Visualization for Strong Story Experience Management Design
This paper presents a software library that enumerates the space of a state transition system specified by an action language, visualizes the states and action connections as a graph, and modifies the visualization based on underlying features determined through state and graph analysis. The library is intended as a tool for strong story interactive narrative design.
more »
« less
- Award ID(s):
- 2303650
- PAR ID:
- 10649696
- Publisher / Repository:
- The AAAI Press
- Date Published:
- Journal Name:
- Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment
- Volume:
- 21
- Issue:
- 1
- ISSN:
- 2326-909X
- Page Range / eLocation ID:
- 399 to 400
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Library learningcompresses a given corpus of programs by extracting common structure from the corpus into reusable library functions. Prior work on library learning suffers from two limitations that prevent it from scaling to larger, more complex inputs. First, it explores too many candidate library functions that are not useful for compression. Second, it is not robust to syntactic variation in the input. We proposelibrary learning modulo theory(LLMT), a new library learning algorithm that additionally takes as input an equational theory for a given problem domain. LLMT uses e-graphs and equality saturation to compactly represent the space of programs equivalent modulo the theory, and uses a novele-graph anti-unificationtechnique to find common patterns in the corpus more directly and efficiently. We implemented LLMT in a tool named babble. Our evaluation shows that babble achieves better compression orders of magnitude faster than the state of the art. We also provide a qualitative evaluation showing that babble learns reusable functions on inputs previously out of reach for library learning.more » « less
-
Human visual understanding of action is reliant on anticipation of contact as is demonstrated by pioneering work in cognitive science. Taking inspiration from this, we introduce representations and models centered on contact, which we then use in action prediction and anticipation. We annotate a subset of the EPIC Kitchens dataset to include time-to-contact between hands and objects, as well as segmentations of hands and objects. Using these annotations we train the Anticipation Module, a module producing Contact Anticipation Maps and Next Active Object Segmentations - novel low-level representations providing temporal and spatial characteristics of anticipated near future action. On top of the Anticipation Module we apply Egocentric Object Manipulation Graphs (Ego-OMG), a framework for action anticipation and prediction. Ego-OMG models longer-term temporal semantic relations through the use of a graph modeling transitions between contact delineated action states. Use of the Anticipation Module within Ego-OMG produces state-of-the-art results, achieving 1st and 2nd place on the unseen and seen test sets, respectively, of the EPIC Kitchens Action Anticipation Challenge, and achieving state-of-the-art results on the tasks of action anticipation and action prediction over EPIC Kitchens. We perform ablation studies over characteristics of the Anticipation Module to evaluate their utility.more » « less
-
Graph-structured data is ubiquitous among a plethora of real-world applications. However, as graph learning algorithms have been increasingly deployed to help decision-making, there has been rising societal concern in the bias these algorithms may exhibit. In certain high-stake decision-making scenarios, the decisions made may be life-changing for the involved individuals. Accordingly, abundant explorations have been made to mitigate the bias for graph learning algorithms in recent years. However, there still lacks a library to collectively consolidate existing debiasing techniques and help practitioners to easily perform bias mitigation for graph learning algorithms. In this paper, we present PyGDebias, an open-source Python library for bias mitigation in graph learning algorithms. As the first comprehensive library of its kind, PyGDebias covers 13 popular debiasing methods under common fairness notions together with 26 commonly used graph datasets. In addition, PyGDebias also comes with comprehensive performance benchmarks and well-documented API designs for both researchers and practitioners. To foster convenient accessibility, PyGDebias is released under a permissive BSD-license together with performance benchmarks, API documentation, and use examples at https://github.com/yushundong/PyGDebias.more » « less
-
Over the past decade, deep reinforcement learning (RL) techniques have significantly advanced robotic systems. However, due to the complex architectures of neural network models, ensuring their trustworthiness is a considerable challenge. Programmatic reinforcement learning has surfaced as a promising approach. Nonetheless, synthesizing robot-control programs remains challenging. Existing methods rely on domain-specific languages (DSLs) populated with user-defined state abstraction predicates and a library of low-level controllers as abstract actions to boot synthesis, which is impractical in unknown environments that lack such predefined components. To address this limitation, we introduce RoboScribe, a novel abstraction refinement-guided program synthesis framework that automatically derives robot state and action abstractions from raw, unsegmented task demonstrations in high-dimensional, continuous spaces. It iteratively enriches and refines an initially coarse abstraction until it generates a task-solving program over the abstracted robot environment. RoboScribe is effective in synthesizing iterative programs by inferring recurring subroutines directly from the robot’s raw, continuous state and action spaces, without needing predefined abstractions. Experimental results show that RoboScribe programs inductively generalize to long-horizon robot tasks involving arbitrary numbers of objects, outperforming baseline methods in terms of both interpretability and efficiency.more » « less
