Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “representation learning”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 757 records · Page 42Linked to original sources
PubMed2000

A model of hippocampally dependent navigation, using the temporal difference learning rule.

This paper presents a model of how hippocampal place cells might be used for spatial navigation in two watermaze tasks: the standard reference memory task and a delayed matching-to-place task. In the reference memory task, the escape platform occupies a single location and rats gradually learn relatively direct paths to the goal over the course of days, in each of which they perform a fixed number of trials. In the delayed matching-to-place task, the escape platform occupies a novel location on each day, and rats gradually acquire one-trial learning, i.e., direct paths on the second trial of each day. The model uses a local, incremental, and statistically efficient connectionist algorithm called temporal difference learning in two distinct components. The first is a reinforcement-based "actor-critic" network that is a general model of classical and instrumental conditioning. In this case, it is applied to navigation, using place cells to provide information about state. By itself, the actor-critic can learn the reference memory task, but this learning is inflexible to changes to the platform location. We argue that one-trial learning in the delayed matching-to-place task demands a goal-independent representation of space. This is provided by the second component of the model: a network that uses temporal difference learning and self-motion information to acquire consistent spatial coordinates in the environment. Each component of the model is necessary at a different stage of the task; the actor-critic provides a way of transferring control to the component that performs best. The model successfully captures gradual acquisition in both tasks, and, in particular, the ultimate development of one-trial learning in the delayed matching-to-place task. Place cells report a form of stable, allocentric information that is well-suited to the various kinds of learning in the model.

Animals↗
PubMed2006

Perspective taking promotes action understanding and learning.

People often learn actions by watching others. The authors propose and test the hypothesis that perspective taking promotes encoding a hierarchical representation of an actor's goals and subgoals-a key process for observational learning. Observers segmented videos of an object assembly task into coarse and fine action units. They described what happened in each unit from either the actor's, their own, or another observer's perspective and later performed the assembly task themselves. Participants who described the task from the actor's perspective encoded actions more hierarchically during observation and learned the task better.

Analysis of Variance↗
PubMed1996

Learning to find your way: a role for the human hippocampal formation.

The importance of the hippocampal formation of the brain for allocentric spatial mapping of the environment has been suggested by animal lesion and electrophysiological work. Here we describe a positron emission tomography (PET) study designed to investigate the regional cerebral blood flow changes associated with topographical memory formation in humans, i.e. the formation of representations of large-scale environments necessary for way-finding. Topographical learning of an urban environment from viewing of film footage depicting navigation was associated with activation of the right parahippocampal gyrus and hippocampus, with activation also of the left parahippocampal gyrus. In addition, there was activity in the pretuneus. In contrast, the encoding of non-navigation episodic memory in a similar realworld context was not associated with activity in the hippocampal formation. Our results shed light on the neural basis of the human representation of large-scale space pinpointing a particular role for the human hippocampal formation in learning to find one's way.

Adult↗
PubMed2002

Does the motor control system use multiple models and context switching to cope with a variable environment?

Studies of arm movements have shown that subjects learn to compensate predictable mechanical perturbations by developing a representation of the relation between the state of motion of the arm and the perturbing forces. Here, we tested the hypothesis that subjects construct internal representations of two different force fields and switch between them when presented with an alternating sequence of these fields. Our results do not support this hypothesis. Subjects performed reaching movements in four sessions over 4 days. On the 1st day the robotic manipulandum perturbed the movement by perpendicular force that alternated its direction after each movement. Subjects were unable to construct the two underlying models and switch between them. On the 2nd day only one field was applied and well learned. On the 3rd day only the other field was applied and well learned. Then the experiment of the 1st day was repeated on the 4th day. Even after this extensive training subjects showed no signs of improved performance with alternating fields. This result combined with previous studies suggests that the central nervous system has a strong tendency to employ a single internal model when dealing with a sequence of perturbations.

Adaptation, Psychological↗
PubMed2001

The brain of musicians. A model for functional and structural adaptation.

Musicians form an ideal subject pool in which one can investigate possible cerebral adaptations to unique requirements of skilled performance as well as cerebral correlates of unique musical abilities such as absolute pitch and others. There are several reasons for this. First, the commencement of musical training usually occurs when the brain and its components may still be able to adapt. Second, musicians undergo long-term motor training and continued practice of complicated bimanual motor activity. Third, imaging studies from our group as well as other groups have shown that motor learning and the acquisition of skills can lead to changes in the representation of motor maps and possibly also to microstructural changes. Whether the unique musical abilities and structural differences that musicians' brains show are due to learning, perhaps during critical periods of brain development and maturation, or whether they reflect innate abilities and capacities that might be fostered by early exposure to music is largely unknown. We will report studies that indicate that certain regions in the brain (corpus callosum, motor cortex, cerebellum) may show some form of adaptation to extraordinary challenges and requirements of performance. These challenges may eventually lead to functional and structural cerebral changes to accommodate the requirements for musical performance. Furthermore, we will also show the neural correlates of one unique musical ability, absolute pitch. This ability may be linked to one structure in the human brain (planum temporale), which is preferentially activated in musicians who have absolute pitch during tone tasks. This structure may undergo some form of functional plasticity that is possible only during a critical period of brain development.

Brain↗
PubMed2005

Semi-automatic learning of simple diagnostic scores utilizing complexity measures.

OBJECTIVE: Knowledge acquisition and maintenance in medical domains with a large application domain ontology is a difficult task. To reduce knowledge elicitation costs, semi-automatic learning methods can be used to support the domain specialists. They are usually not only interested in the accuracy of the learned knowledge: the understandability and interpretability of the learned models is of prime importance as well. Then, often simple models are more favorable than complex ones. METHODS AND MATERIAL: We propose diagnostic scores as a promising approach for the representation of simple diagnostic knowledge, and present a method for inductive learning of diagnostic scores. It can be incrementally refined by including background knowledge. We present complexity measures for determining the complexity of the learned scores. RESULTS: We give an evaluation of the presented approach using a case base from the fielded system SonoConsult. We further discuss that the user can easily balance between accuracy and complexity of the learned knowledge applying the presented measures. CONCLUSIONS: We argue that semi-automatic learning methods can support the domain specialist efficiently when building (diagnostic) knowledge systems from scratch. The presented complexity measures allow for an intuitive assessment of the learned patterns.

Algorithms↗
PubMed2005

In search of the motor engram: motor map plasticity as a mechanism for encoding motor experience.

Motor skill acquisition occurs through modification and organization of muscle synergies into effective movement sequences. The learning process is reflected neurophysiologically as a reorganization of movement representations within the primary motor cortex, suggesting that the motor map is a motor engram. However, the specific neural mechanisms underlying map plasticity are unknown. Here the authors review evidence that 1) motor map topography reflects the capacity for skilled movement, 2) motor skill learning induces reorganization of motor maps in a manner that reflects the kinematics of acquired skilled movement, 3) map plasticity is supported by a reorganization of cortical microcircuitry involving changes in synaptic efficacy, and 4) motor map integrity and topography are influenced by various neurochemical signals that coordinate changes in cortical circuitry to encode motor experience. Finally, the role of motor map plasticity in recovery of motor function after brain damage is discussed.

Animals↗
PubMed2003

Hippocampal lesions and discrimination performance of mice in the radial maze: sparing or impairment depending on the representational demands of the task.

The effects of ibotenate hippocampal lesions on discrimination performance in an eight-arm radial maze were investigated in mice, using a three-stage paradigm in which the only parameter that varied among stages was the way the arms were presented. In the initial learning phase (stage 1), animals learned the valence or reward contingency associated with six (three positive and three negative) adjacent arms of the maze using a successive (go/no-go) discrimination procedure. In the first test phase (stage 2), the six arms were grouped into three pairs, so that on each trial, the subject was faced with a choice between two adjacent arms of opposite valence (concurrent two-choice discrimination). In the second test phase (stage 3), the subject was faced with all six arms simultaneously (six-choice discrimination). Hippocampal-lesioned mice acquired the initial learning phase at a near-normal rate but behaved as if they had learned nothing when challenged with the two-choice discriminations at stage 2. In contrast, they behaved normally when confronted with the six-choice discrimination at stage 3. Detailed examination of within- and between-stage performance suggests that hippocampal-lesioned mice perform as intact mice when presentation of the discriminanda encourages the storage and use of separate representations (i.e., in initial learning and six-choice discrimination testing), but that they fail in test situations that involve explicit comparisons between such separate representations (two-choice discriminations), hence requiring the use of relational representations.

Algorithms↗
PubMedDate not supplied

A model of STDP based on spatially and temporally local information: derivation and combination with gated decay.

Temporal relationships between neuronal firing and plasticity have received significant attention in recent decades. Neurophysiological studies have shown the phenomenon of spike-timing-dependent plasticity (STDP). Various models were suggested to implement an STDP-like learning rule in artificial networks based on spiking neuronal representations. The rule presented here was developed under three constraints. First, it only depends on the information that is available at the synapse at the time of synaptic modification. Second, it naturally follows from neurophysiological and psychological research starting with Hebb's postulate [D. Hebb. (1949). The organization of behavior. Wiley, New York]. Third, it is simple, computationally cheap and its parameters are straightforward to determine. This rule is further extended by addition of four different types of gating derived from conventionally used types of gated decay in learning rules for continuous firing rate neural networks. The results show that the advantages of using these gatings are transferred to the new rule without sacrificing its dependency on spike-timing.

Algorithms↗
PubMed2003

Enhanced cholinergic suppression of previously strengthened synapses enables the formation of self-organized representations in olfactory cortex.

Computational modeling assists in analyzing the specific functional role of the cellular effects of acetylcholine within cortical structures. In particular, acetylcholine may regulate the dynamics of encoding and retrieval of information by regulating the magnitude of synaptic transmission at excitatory recurrent connections. Many abstract models of associative memory function ignore the influence of changes in synaptic strength during the storage process and apply the effect of these changes only during a so-called recall-phase. Efforts to ensure stable activity with more realistic, continuous updating of the synaptic strength during the storage process have shown that the memory capacity of a realistic cortical network can be greatly enhanced if cholinergic modulation blocks transmission at synaptic connections of the association fibers during the learning process. We here present experimental data from an olfactory cortex brain slice preparation showing that previously potentiated fibers show significantly greater suppression (presynaptic inhibition) by the cholinergic agonist carbachol than unpotentiated fibers. We conclude that low suppression of non-potentiated fibers during the learning process ensures the formation of self-organized representations in the neural network while the higher suppression of previously potentiated fibers minimizes interference between overlapping patterns. We show in a computational model of olfactory cortex, that, together, these two phenomena reduce the overlap between patterns that are stored within the same neural network structure. These results further demonstrate the contribution of acetylcholine to mechanisms of cortical plasticity. The results are consistent with the extensive evidence supporting a role for acetylcholine in encoding of new memories and enhancement of response to salient sensory stimuli.

Carbachol↗
PubMed1999

Modular features of motor control and learning.

The study of complex motor behaviours has highlighted the role of modular representations both in the planning and in the execution of actions. Recent findings suggest the presence of functional modules within a variety of neural structures. Computational investigations are now addressing the issue of how these modules may act concurrently to generate a wide repertoire of behaviours.

Animals↗
PubMed2004

The role of cognition in classical and operant conditioning.

For the past 35 years, learning theorists have been providing models that depend on mental representations, even in their most simple, deterministic, and mechanistic approaches. Hence, cognitive involvement (typically thought of as expectancy) is assumed for most instances of classical and operant conditioning, with current theoretical differences concerning the level of cognition that is involved (e.g., simple association vs. rule learning), rather than its presence. Nevertheless, many psychologists not in the mainstream of learning theory continue to think of cognitive and conditioning theories as rival families of hypotheses. In this article, the data pertaining to the role of higher-order cognition in conditioning is reviewed, and a theoretical synthesis is proposed that provides a role for both automatic and cognitively mediated processes.

Cognition↗
PubMed1984

Semantic and visual memory codes in learning disabled readers.

Two experiments investigated whether learning disabled readers' impaired recall is due to multiple coding deficiencies. In Experiment 1, learning disabled and skilled readers viewed nonsense pictures without names or with either relevant or irrelevant names with respect to the distinctive characteristics of the picture. Both types of names improved recall of nondisabled readers, while learning disabled readers exhibited better recall for unnamed pictures. No significant difference in recall was found between name training (relevant, irrelevant) conditions within reading groups. In Experiment 2, both reading groups participated in recall training for complex visual forms labeled with unrelated words, hierarchically related words, or without labels. A subsequent reproduction transfer task showed a facilitation in performance in skilled readers due to labeling, with learning disabled readers exhibiting better reproduction for unnamed pictures. Measures of output organization (clustering) indicated that recall is related to the development of superordinate categories. The results suggest that learning disabled children's reading difficulties are due to an inability to activate a semantic representation that interconnects visual and verbal codes.

Adolescent↗
PubMed2006

Learning words and rules: abstract knowledge of word order in early sentence comprehension.

Children quickly acquire basic grammatical facts about their native language. Does this early syntactic knowledge involve knowledge of words or rules? According to lexical accounts of acquisition, abstract syntactic and semantic categories are not primitive to the language-acquisition system; thus, early language comprehension and production are based on verb-specific knowledge. The present experiments challenge this account: We probed the abstractness of young children's knowledge of syntax by testing whether 25- and 21-month-olds extend their knowledge of English word order to new verbs. In four experiments, children used word order appropriately to interpret transitive sentences containing novel verbs. These findings demonstrate that although toddlers have much to learn about their native languages, they represent language experience in terms of an abstract mental vocabulary. These abstract representations allow children to rapidly detect general patterns in their native language, and thus to learn rules as well as words from the start.

Child Language↗
PubMed1988

Representation of serial order in monkeys (Cebus apella).

Cebus monkeys were trained on a five-item serial learning task, symbolized as ABCDE; the initial stages of training were on the shorter subseries AB, ABC, and ABCD. To assess the monkeys' knowledge of the sequential position of each item, pair-wise tests were given to 2 subjects after acquisition of the ABCD series and to 4 subjects after reaching criterion on the ABCDE series. In both tests, the monkeys performed at high levels on the interior pairs, which were BC for the ABCD series, and BC, BD, and CD for the ABCDE series. These results, as well as the orderly relations observed in the pair-wise tests between first-item response latency and first-item position and between second-item response latency and number of missing items, indicated that the monkeys had developed a well-organized internal representation of the four- and five-item series. Although pigeons are also capable of learning four-item and five-item series, they apparently do not develop a comparable representational structure. The disparity between the monkeys' and pigeons' representational competence for serial order is predictable from the difference in their capacities for associative transitivity.

Animals↗
PubMed2006

Formation of category representations.

Many formal models of categorization assume, implicitly or explicitly, that categorization results in the formation of direct associations from representations of the presented stimuli to representations of the experimentally provided category labels. In three categorization experiments employing a polymorphous classification structure (Dennis, Hampton, & Lea, 1973) and a partial reversal, optional shift procedure (Kendler, Kendler, & Wells, 1960), we provide evidence consistent with the hypothesis that learning a new classification problem results in the creation of category representations that mediate between representations of the stimulus and the label. This hypothesis can be instantiated through the AMBRY model (Kruschke, 1996).

Adult↗
PubMed1995

Steering through the murky waters of a scientific conflict: situated and symbolic models of clinical cognition.

The situated action perspective, which embraces a diversity of views, challenges several of the fundamental assumptions of the symbolic information-processing framework underlying cognitive science and artificial intelligence. In this paper, we consider the following issues; symbolic representations, plans and actions, distributed cognition, and the transfer of learning. We evaluate each of these issues in terms of research and theories in clinical cognition and examine the implications for education and training, and for the integration of intelligent systems in medical practice. We argue for a reconceptualization of the symbolic framework in terms of the way the role of internal representations and cognitive activities are perceived. However, symbolic representations are integral to medical cognition and should continue to be central in any theoretical framework. A re-examination of cognitive science in medicine in terms of the relationship among physicians, technology, and the workplace could prove to be constructive in bridging the gap between theory and practice.

Artificial Intelligence↗
PubMed2001

Menu design with visual momentum for compact smart products.

Users of compact smart products with small screens often have trouble learning the menu structure. If they cannot master the menu structure, users are not able to fully utilize the products. It is argued in this paper that using visual momentum in menu representation design helps users develop effective mental maps of menu structures and promotes learning of the user interface. To assess the effect of visual momentum in this study, four types of menu representations were developed. Additionally, two menu hierarchies, two types of function key layout, and two types of function key labeling were assessed to examine the effects of menu dimension and compatibility. Experimental results indicated that participants using a partial menu map with visual momentum design performed the best, and participants using a partial menu map without visual momentum performed the poorest, even worse than those-using command-only representation. The results also showed that the menu navigation problem appeared to be particularly significant with a deep menu hierarchy. Actual or potential applications of this research include menu representation design for compact smart products.

Adult↗