Representation
A representation is in a sense a vector of features or a symbol which (ideally) uniquely correspond to an object (i.e. a noun) and properties of said object (verbs, descriptors, etc…). Many modern interpretations of a representation are often traced back to Charles S. Peirce’s Semiotics.
The Representational Theory of Mind
This theory of mind takes mental states (i.e. internal beliefs, perceptions, thoughts, etc..) as a basis. These states are said to have intentionality, which means they refer to things (i.e. representations).
This theory then understands cognition as sequences of representations (intentional mental states).
Links to this note:
- Current Learning Objectives
- white2015developing: Developing a predictive approach to knowledge
- synofzik2013experience: The experience of agency: an interplay between prediction and postdiction
- sutton2011horde: Horde: A Scalable Real-time Architecture for Learning Knowledge from Unsupervised Sensorimotor Interaction
- sternberg2016cognitive: Cognitive Psychology
- Helmholtz Sign Theory
- scholkopf2019causality: Causality for Machine Learning
- schwarzer2021pretraining: Pretraining Representations for Data-Efficient Reinforcement Learning
- roy2018editorial: Editorial: Representation in the Brain
- Reinforcement Learning: An Introduction
- Puddle World
- prinz1990common: A Common Coding Approach to Perception and Action
- Predictive Processing
- Peirce Semiotic
- Offline RL (Modl)
- Mountain Car
- Model-based RL
- littman2002predictive: Predictive Representations of State
- lehnert2017advantages: Advantages and Limitations of using Successor Features for Transfer in Reinforcement Learning
- kostas2019asynchronous: Asynchronous Coagent Networks: Stochastic Networks for Reinforcement Learning without Backpropagation or a Clock
- kearney2019making: Making Meaning: Semiotics Within Predictive Knowledge Architectures
- Interview Review Material
- Hypothesis
- Hermann von Helmholtz
- Dimensionality Reduction
- dayan1993improving: Improving Generalization for Temporal Difference Learning: The Successor Representation
- clark2013whatever: Whatever next? Predictive brains, situated agents, and the future of cognitive science
- chandar2019nonsaturating: Towards Non-saturating Recurrent Units for Modelling Long-term Dependencies
- Causality
- Calculus
- bubic2010prediction: Prediction, cognition and the brain
- Auxiliary Tasks
- Why am I blogging?
- About