Results 1  10
of
194
Locally weighted learning
 ARTIFICIAL INTELLIGENCE REVIEW
, 1997
"... This paper surveys locally weighted learning, a form of lazy learning and memorybased learning, and focuses on locally weighted linear regression. The survey discusses distance functions, smoothing parameters, weighting functions, local model structures, regularization of the estimates and bias, ass ..."
Abstract

Cited by 599 (51 self)
 Add to MetaCart
This paper surveys locally weighted learning, a form of lazy learning and memorybased learning, and focuses on locally weighted linear regression. The survey discusses distance functions, smoothing parameters, weighting functions, local model structures, regularization of the estimates and bias, assessing predictions, handling noisy data and outliers, improving the quality of predictions by tuning t parameters, interference between old and new data, implementing locally weighted learning e ciently, and applications of locally weighted learning. A companion paper surveys how locally weighted learning can be used in robot learning and control.
The Spatial Semantic Hierarchy
 Artificial Intelligence
, 2000
"... The Spatial Semantic Hierarchy is a model of knowledge of largescale space consisting of multiple interacting representations, both qualitative and quantitative. The SSH is inspired by the properties of the human cognitive map, and is intended to serve both as a model of the human cognitive map and ..."
Abstract

Cited by 339 (35 self)
 Add to MetaCart
(Show Context)
The Spatial Semantic Hierarchy is a model of knowledge of largescale space consisting of multiple interacting representations, both qualitative and quantitative. The SSH is inspired by the properties of the human cognitive map, and is intended to serve both as a model of the human cognitive map and as a method for robot exploration and mapbuilding. The multiple levels of the SSH express states of partial knowledge, and thus enable the human or robotic agent to deal robustly with uncertainty during both learning and problemsolving. The control level represents useful patterns of sensorimotor interaction with the world in the form of trajectoryfollowing and hillclimbing control laws leading to locally distinctive states. Local geometric maps in local frames of reference can be constructed at the control level to serve as observers for control laws in particular neighborhoods. The causal level abstracts continuous behavior among distinctive states into a discrete model ...
Constructive Incremental Learning From Only Local Information
 NEURAL COMPUTATION
"... We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of the receptive field of each locally linear model as well as the parameters of the locally linear mod ..."
Abstract

Cited by 208 (40 self)
 Add to MetaCart
We introduce a constructive, incremental learning system for regression problems that models data by means of spatially localized linear models. In contrast to other approaches, the size and shape of the receptive field of each locally linear model as well as the parameters of the locally linear model itself are learned independently, i.e., without the need for competition or any other kind of communication. Independent learning is accomplished by incrementally minimizing a weighted local cross validation error. As a result, we obtain a learning system that can allocate resources as needed while dealing with the biasvariance dilemma in a principled way. The spatial localization of the linear models increases robustness towards negative interference. Our learning system can be interpreted as a nonparametric adaptive bandwidth smoother, as a mixture of experts where the experts are trained in isolation, and as a learning system which profits from combining independent expert knowledge on the same problem. This paper illustrates the potential learning capabilities of purely local learning and offers an interesting and powerful approach to learning with receptive fields.
Stochastic Dynamic Programming with Factored Representations
, 1997
"... Markov decision processes(MDPs) have proven to be popular models for decisiontheoretic planning, but standard dynamic programming algorithms for solving MDPs rely on explicit, statebased specifications and computations. To alleviate the combinatorial problems associated with such methods, we prop ..."
Abstract

Cited by 189 (10 self)
 Add to MetaCart
(Show Context)
Markov decision processes(MDPs) have proven to be popular models for decisiontheoretic planning, but standard dynamic programming algorithms for solving MDPs rely on explicit, statebased specifications and computations. To alleviate the combinatorial problems associated with such methods, we propose new representational and computational techniques for MDPs that exploit certain types of problem structure. We use dynamic Bayesian networks (with decision trees representing the local families of conditional probability distributions) to represent stochastic actions in an MDP, together with a decisiontree representation of rewards. Based on this representation, we develop versions of standard dynamic programming algorithms that directly manipulate decisiontree representations of policies and value functions. This generally obviates the need for statebystate computation, aggregating states at the leaves of these trees and requiring computations only for each aggregate state. The key to these algorithms is a decisiontheoretic generalization of classic regression analysis, in which we determine the features relevant to predicting expected value. We demonstrate the method empirically on several planning problems,
A Review and Empirical Evaluation of Feature Weighting Methods for a Class of Lazy Learning Algorithms
 ARTIFICIAL INTELLIGENCE REVIEW
, 1997
"... Many lazy learning algorithms are derivatives of the knearest neighbor (kNN) classifier, which uses a distance function to generate predictions from stored instances. Several studies have shown that kNN's performance is highly sensitive to the definition of its distance function. Many k ..."
Abstract

Cited by 147 (0 self)
 Add to MetaCart
(Show Context)
Many lazy learning algorithms are derivatives of the knearest neighbor (kNN) classifier, which uses a distance function to generate predictions from stored instances. Several studies have shown that kNN's performance is highly sensitive to the definition of its distance function. Many kNN variants have been proposed to reduce this sensitivity by parameterizing the distance function with feature weights. However, these variants have not been categorized nor empirically compared. This paper reviews a class of weightsetting methods for lazy learning algorithms. We introduce a framework for distinguishing these methods and empirically compare them. We observed four trends from our experiments and conducted further studies to highlight them. Our results suggest that methods which use performance feedback to assign weight settings demonstrated three advantages over other methods: they require less preprocessing, perform better in the presence of interacting features, and generally require less training data to learn good settings. We also found that continuous weighting methods tend to outperform feature selection algorithms for tasks where some features are useful but less important than others.
An introduction to collective intelligence
 Handbook of Agent technology. AAAI
, 1999
"... ..."
(Show Context)
Performance Animation from Lowdimensional Control Signals
 ACM Transactions on Graphics
, 2005
"... This paper introduces an approach to performance animation that employs video cameras and a small set of retroreflective markers to create a lowcost, easytouse system that might someday be practical for home use. The lowdimensional control signals from the user's performance are supplement ..."
Abstract

Cited by 129 (18 self)
 Add to MetaCart
This paper introduces an approach to performance animation that employs video cameras and a small set of retroreflective markers to create a lowcost, easytouse system that might someday be practical for home use. The lowdimensional control signals from the user's performance are supplemented by a database of prerecorded human motion. At run time, the system automatically learns a series of local models from a set of motion capture examples that are a close match to the marker locations captured by the cameras. These local models are then used to reconstruct the motion of the user as a fullbody animation. We demonstrate the power of this approach with realtime control of six different behaviors using two video cameras and a small set of retroreflective markers. We compare the resulting animation to animation from commercial motion capture equipment with a full set of markers.
Monte Carlo POMDPs
"... We present a Monte Carlo algorithm for learning to act in partially observable Markov decision processes (POMDPs) with realvalued state and action spaces. Our approach uses importance sampling for representing beliefs, and Monte Carlo approximation for belief propagation. A reinforcement learning a ..."
Abstract

Cited by 119 (4 self)
 Add to MetaCart
We present a Monte Carlo algorithm for learning to act in partially observable Markov decision processes (POMDPs) with realvalued state and action spaces. Our approach uses importance sampling for representing beliefs, and Monte Carlo approximation for belief propagation. A reinforcement learning algorithm, value iteration,is employed to learn value functions over belief states. Finally, a samplebased version of nearest neighbor is used to generalize across states. Initial empirical results suggest that our approach works well in practical applications.
Precomputing interactive dynamic deformable scenes
 ACM Trans. Graph
, 2003
"... dynamics by driving the scene with parameterized interactions representative of runtime usage. (b) Model reduction on observed dynamic deformations yields a lowrank approximation to the system’s parameterized impulse response functions. (c) Deformed state geometries are then sampled and used to pre ..."
Abstract

Cited by 90 (8 self)
 Add to MetaCart
dynamics by driving the scene with parameterized interactions representative of runtime usage. (b) Model reduction on observed dynamic deformations yields a lowrank approximation to the system’s parameterized impulse response functions. (c) Deformed state geometries are then sampled and used to precompute and coparameterize a radiance transfer model for deformable objects. (d) The final simulation responds plausibly to interactions similar to those precomputed, includes complex collision and global illumination effects, and runs in real time. We present an approach for precomputing datadriven models of interactive physically based deformable scenes. The method permits realtime hardware synthesis of nonlinear deformation dynamics, including selfcontact and global illumination effects, and supports realtime user interaction. We use datadriven tabulation of the system’s deterministic state space dynamics, and model reduction to build efficient lowrank parameterizations of the deformed shapes. To support runtime interaction, we also tabulate impulse response functions for a palette of external excitations. Although our approach simulates particular systems under very particular interaction conditions, it has several advantages. First, parameterizing all possible scene deformations enables us to precompute novel reduced coparameterizations of global scene illumination for lowfrequency lighting conditions. Second, because the deformation dynamics are precomputed and parameterized as a whole, collisions are resolved within the scene during precomputation so that runtime selfcollision handling is implicit. Optionally, the datadriven models can be synthesized on programmable graphics hardware, leaving only the lowdimensional state space dynamics and appearance data models to be computed by the main CPU.
Intelligence by Design: Principles of Modularity and Coordination for Engineering Complex Adaptive Agents
, 2001
"... All intelligence relies on search  for example, the search for an intelligent agent's next action. Search is only likely to succeed in resourcebounded agents if they have already been biased towards finding the right answer. In artificial agents, the primary source of bias is engineering. T ..."
Abstract

Cited by 81 (27 self)
 Add to MetaCart
All intelligence relies on search  for example, the search for an intelligent agent's next action. Search is only likely to succeed in resourcebounded agents if they have already been biased towards finding the right answer. In artificial agents, the primary source of bias is engineering. This dissertation