Human-inspired robotic forgetting: Filtering to improve estimation accuracy

Article · January 2009with 32 Reads
Abstract
Perfect memory and recall provides a mixed blessing. While flawless recollection of episodic data allows for increased reasoning, photographic memory can hinder a robot's ability to operate in real-time dynamic environ-ments. Human-inspired forgetting methods may enable robotic systems to rid themselves of out-dated, irrelevant, and erroneous data. This paper presents the ActSimple al-gorithm and an associated experimental analysis. The Act-Simple algorithm is a novel approach to improving robotic performance by filtering data available to existing algo-rithms. The experimental analysis tested the effectiveness of five forgetting algorithms in a WiFi signal strength es-timation task. The results suggest that forgetting can im-prove estimation accuracy while reducing the number of sensor readings required. The simplified version of Act-Simple outperformed the other forgetting methods and ap-pears to be a flexible and adaptable means of incorporating human-inspired forgetting into robotic systems.

Do you want to read the rest of this article?

Request full-text
Request Full-text Paper PDF
  • Conference Paper
    Full-text available
    To interact naturally with humans, robots need to be aware of their own surroundings. This awareness is usually encoded in some implicit or explicit representation of the situated context. In this paper, we present a new framework for constructing rich belief models of the robot's environment. Key to our approach is the use of Markov Logic as a unified framework for inference over these beliefs. Markov Logic is a combination of first-order logic and probabilistic graphical models. Its expressive power allows us to capture both the rich relational structure of the environment and the uncertainty arising from the noise and incompleteness of low-level sensory data. The constructed belief models evolve dynamically over time and incorporate various contextual information such as spatio-temporal framing, multi-agent epistemic status, and saliency measures. Beliefs can also be referenced and extended “top-down” via linguistic communication. The approach is being integrated into a cognitive architecture for mobile robots interacting with humans using spoken dialogue.
  • Article
    Robots are frequently presented with vast arrays of diverse data. Unfortunately, perfect memory and recall provides a mixed blessing. While flawless recollection of episodic data allows increased reasoning, photographic memory can hinder a robot's ability to operate in real-time dynamic environments. Human-inspired forgetting methods may enable robotic systems to rid themselves of out-dated, irrelevant, and erroneous data. This paper presents the use of human-inspired forgetting to act as a filter, removing unnecessary, erroneous, and out-of-date information. The novel ActSimple forgetting algorithm has been developed specifically to provide effective forgetting capabilities to robotic systems. This paper presents the ActSimple algorithm and how it was optimized and tested in a WiFi signal strength estimation task. The results generated by real-world testing suggest that human-inspired forgetting is an effective means of improving the ability of mobile robots to move and operate within complex and dynamic environments.
  • Article
    Full-text available
    Psychologists often assume that short-term storage is synonymous with activation, a mnemonic property that keeps information in an immediately accessible form. Permanent knowledge is activated, as a result of on-line cognitive processing, and an activity trace is established "in" short-term (or working) memory. Activation is assumed to decay spontaneously with the passage of time, so a refreshing process-rehearsal-is needed to maintain availability. Most of the phenomena of immediate retention, such as capacity limitations and word length effects, are assumed to arise from trade-offs between rehearsal and decay. This "standard model" of how we remember over the short-term still enjoys considerable popularity, although recent research questions most of its main assumptions. In this chapter I review the recent research and identify the empirical and conceptual problems that plague traditional conceptions of short-term memory. Increasingly, researchers are recognizing that short-term retention is cue driven, much like long-term memory, and that neither rehearsal nor decay is likely to explain the particulars of short-term forgetting.
  • Article
    In many fields using empirical areal data there arises a need for interpolating from irregularly-spaced data to produce a continuous surface. These irregularly-spaced locations, hence referred to as “data points,” may have diverse meanings: in meterology, weather observation stations; in geography, surveyed locations; in city and regional planning, centers of data-collection zones; in biology, observation locations. It is assumed that a unique number (such as rainfall in meteorology, or altitude in geography) is associated with each data point. In order to display these data in some type of contour map or perspective view, to compare them with data for the same region based on other data points, or to analyze them for extremes, gradients, or other purposes, it is extremely useful, if not essential, to define a continuous function fitting the given values exactly. Interpolated values over a fine grid may then be evaluated. In using such a function it is assumed that the original data are without error, or that compensation for error will be made after interpolation.
  • Article
    In this paper, we present successful strategies for forgetting cases in a Case-Based Reasoning (CBR) system applied to autonomous robot navigation. This extends previous work that involved a CBR architecture which indexes cases by the spatio-temporal characteristics of the sensor data, and outputs or selects parameters of behaviors in a behavior-based robot architecture. In such a system, the removal of cases can be applied when a new situation unlike any current case in the library is encountered, but the library is full. Various strategies of determining which cases to remove are proposed, including metrics such as how frequently a case is used and a novel spreading activation mechanism. Experimental results show that such mechanisms can increase the performance of the system significantly and allow it to essentially forget old environments in which it was trained in favor of new environments it is currently encountering. The performance of this new system is better than both a purely reactive behavior-based system as well as the CBR module that did not forget cases. Furthermore, such forgetting mechanisms can be useful even when there is no major environmental shift during training, since some cases can potentially be harmful or rarely used. The relationship between the forgetting mechanism and the case library size is also discussed.
  • Article
    Functional decay theory proposes that decay and interference, historically viewed as competing accounts of forgetting, are instead functionally related. The theory posits that (a) when an attribute must be updated frequently in memory, its current value decays to prevent interference with later values, and (b) the decay rate adapts to the rate of memory updates. Behavioral predictions of the theory were tested in a task-switching paradigm in which memory for the current task had to be updated every few seconds, hundreds of times. Reaction times and error rates both increased gradually between updates, reflecting decay of memory for the current task. This performance decline was slower when updates were less frequent, reflecting a decrease in the decay rate following a decrease in the update rate. A candidate mechanism for controlled decay is proposed, the data are reconciled with practice effects, and implications for models of executive control are discussed.
  • Article
    Full-text available
    Traditional theories of forgetting are wedded to the notion that cue-overload interference procedures (often involving the A-B, A-C list-learning paradigm) capture the most important elements of forgetting in everyday life. However, findings from a century of work in psychology, psychopharmacology, and neuroscience converge on the notion that such procedures may pertain mainly to forgetting in the laboratory and that everyday forgetting is attributable to an altogether different form of interference. According to this idea, recently formed memories that have not yet had a chance to consolidate are vulnerable to the interfering force of mental activity and memory formation (even if the interfering activity is not similar to the previously learned material). This account helps to explain why sleep, alcohol, and benzodiazepines all improve memory for a recently learned list, and it is consistent with recent work on the variables that affect the induction and maintenance of long-term potentiation in the hippocampus.
  • Article
    Full-text available
    Time-based theories expect memory performance to decline as the delay between study and recall of an item increases. The assumption of time-based forgetting, central to many models of serial recall, underpins their key behaviors. Here we compare the predictions of time-based and event-based models by simulation and test them in two experiments using a novel manipulation of the delay between study and retrieval. Participants were trained, via corrective feedback, to recall at different speeds, thus varying total recall time from 6 to 10 sec. In the first experiment, participants used the keyboard to enter their responses but had to repeat a word (called the suppressor) aloud during recall to prevent rehearsal. In the second experiment, articulation was again required, but recall was verbal and was paced by the number of repetitions of the suppressor in between retrieval of items. In both experiments, serial position curves for all retrieval speeds overlapped, and output time had little or no effect. Comparative evaluation of a time-based and an event-based model confirmed that these results present a particular challenge to time-based approaches. We conclude that output interference, rather than output time, is critical in serial recall.
  • Article
    Full-text available
    A model of memory retrieval is described. The model embodies 4 main claims: (a) temporal memory--traces of items are represented in memory partly in terms of their temporal distance from the present; (b) scale-similarity--similar mechanisms govern retrieval from memory over many different timescales; (c) local distinctiveness--performance on a range of memory tasks is determined by interference from near psychological neighbors; and (d) interference-based forgetting--all memory loss is due to interference and not trace decay. The model is applied to data on free recall and serial recall. The account emphasizes qualitative similarity in the retrieval principles involved in memory performance at all timescales, contrary to models that emphasize distinctions between short-term and long-term memory.
  • Article
    The past 10 years have brought near-revolutionary changes in psychological theories about short-term memory, with similarly great advances in the neurosciences. Here, we critically examine the major psychological theories (the "mind") of short-term memory and how they relate to evidence about underlying brain mechanisms. We focus on three features that must be addressed by any satisfactory theory of short-term memory. First, we examine the evidence for the architecture of short-term memory, with special attention to questions of capacity and how--or whether--short-term memory can be separated from long-term memory. Second, we ask how the components of that architecture enact processes of encoding, maintenance, and retrieval. Third, we describe the debate over the reason about forgetting from short-term memory, whether interference or decay is the cause. We close with a conceptual model tracing the representation of a single item through a short-term memory task, describing the biological mechanisms that might support psychological processes on a moment-by-moment basis as an item is encoded, maintained over a delay with some forgetting, and ultimately retrieved.
  • Article
    This paper is a discussion of the relationship between learning and forgetting. An analysis of the economics of learning is carried out and it is argued that knowledge can sometimes have a negative value. A series of experiments involving a program which learns to traverse state spaces is described. It is shown that most of the knowledge acquired is of negative value even though it is correct and was acquired solving similar problems. It is shown that the value of the knowledge depends on what else is known and that random forgetting can sometimes lead to substantial improvements in performance. It is concluded that research into knowledge acquisition should take seriously the possibility that knowledge may sometimes be harmful. The view is taken that learning and forgetting are complementary processes which construct and maintain useful representations of experience. 1. Introduction Research on machine learning is concerned with the problem of how a system may acquire knowledge that...
  • Forgetting to remem-ber: Thefunctionalrelationshipofdecayandinterference
    • E M Altmann
    E.M. Altmann and W.D. Gray, Forgetting to remem-ber: Thefunctionalrelationshipofdecayandinterference, Psychological Science, 13, 2002, 27-33.
  • The psychology and neuroscience of forgetting Forgetting to remember: The functional relationship of decay and interference
    • J T Wixted
    J.T. Wixted, The psychology and neuroscience of forgetting, Annual Review of Psychology, 55, 2004, 235-269. [7] E.M. Altmann and W.D. Gray, Forgetting to remember: The functional relationship of decay and interference, Psychological Science, 13, 2002, 27-33.
  • Human inspired forgetting for robotics systems
    • S T Freedman
    S.T. Freedman, Human inspired forgetting for robotics systems: Ph.D. Proposal, Tech. Rep. HMT-09-02, Human-Machine Laboratory, 2009.