Artificial General Intelligence

Frontmatter

Hybrid Strategies Towards Safe “Self-Aware” Superintelligent Systems

Abstract

Against the backdrop of increasing progresses in AI research paired with a rise of AI applications in decision-making processes, security-critical domains as well as in ethically relevant frames, a large-scale debate on possible safety measures encompassing corresponding long-term and short-term issues has emerged across different disciplines. One pertinent topic in this context which has been addressed by various AI Safety researchers is e.g. the AI alignment problem for which no final consensus has been achieved yet. In this paper, we present a multidisciplinary toolkit of AI Safety strategies combining considerations from AI and Systems Engineering as well as from Cognitive Science with a security mindset as often relevant in Cybersecurity. We elaborate on how AGI “Self-awareness” could complement different AI Safety measures in a framework extended by a jointly performed Human Enhancement procedure. Our analysis suggests that this hybrid framework could contribute to undertake the AI alignment problem from a new holistic perspective through security-building synergetic effects emerging thereof and could help to increase the odds of a possible safe future transition towards superintelligent systems.

Nadisha-Marie Aliman, Leon Kester

Request Confirmation Networks in MicroPsi 2

Abstract

To combine neural learning with the sequential detection of hierarchies of sensory features, and to facilitate planning and script execution, we propose Request Confirmation Networks (ReCoNs). ReCoNs are spreading activation networks with units that contain an activation and a state, and are connected by typed directed links that indicate partonomic relations and spatial or temporal succession. By passing activation along the links, ReCoNs can perform both neural computations and controlled script execution. We demonstrate the application of ReCoNs in the context of performing simple arithmetic, based on camera images of mathematical expressions.

Joscha Bach, Katherine Gallagher

Task Analysis for Teaching Cumulative Learners

Abstract

A generally intelligent machine (AGI) should be able to learn a wide range of tasks. Knowledge acquisition in complex and dynamic task-environments cannot happen all-at-once, and AGI-aspiring systems must thus be capable of cumulative learning: efficiently making use of existing knowledge during learning, supporting increases in the scope of ability and knowledge, incrementally and predictably — without catastrophic forgetting or mangling of existing knowledge. Where relevant expertise is at hand the learning process can be aided by curriculum-based teaching, where a teacher divides a high-level task up into smaller and simpler pieces and presents them in an order that facilitates learning. Creating such a curriculum can benefit from expert knowledge of (a) the task domain, (b) the learning system itself, and (c) general teaching principles. Curriculum design for AI systems has so far been rather ad-hoc and limited to systems incapable of cumulative learning. We present a task analysis methodology that utilizes expert knowledge and is intended to inform the construction of teaching curricula for cumulative learners. Inspired in part by methods from knowledge engineering and functional requirements analysis, our strategy decomposes high-level tasks in three ways based on involved actions, features and functionality. We show how this methodology can be used for a (simplified) arrival control task from the air traffic control domain, where extensive expert knowledge is available and teaching cumulative learners is required to facilitate the safe and trustworthy automation of complex workflows.

Jordi E. Bieger, Kristinn R. Thórisson

Associative Memory: An Spiking Neural Network Robotic Implementation

Abstract

This article proposes a novel minimalist bio-inspired associative memory (AM) mechanism based on a spiking neural network acting as a controller in simple virtual and physical robots. As such, several main features of a general AM concept were reproduced. Using the strength of temporal coding at the single spike resolution level, this study approaches the AM phenomenon with basic examples in the visual modality. Specifically, the AM include varying time delays in synaptic links and asymmetry in the spike-timing dependent plasticity learning rules to solve visual tasks of pattern-matching, pattern-completion and noise-tolerance for autoassociative and heteroassociative memories. This preliminary work could serve as a step toward future comparative analysis with traditional artificial neural networks.

André Cyr, Frédéric Thériault, Matthew Ross, Sylvain Chartier

A Comprehensive Ethical Framework for AI Entities: Foundations

Abstract

The participation of AI in society is expected to increase significantly, and with that the scope, intensity and significance of morally-burdened effects produced or otherwise related to AI, and the possible future advent of AGI. There is a lack of a comprehensive ethical framework for AI and AGI, which can help manage moral scenarios in which artificial entities are participants. Therefore, I propose the foundations of such a framework in this text, and suggest that it can enable artificial entities to make morally sound decisions in complex moral scenarios.

Andrej Dameski

Partial Operator Induction with Beta Distributions

Abstract

A specialization of Solomonoff Operator Induction considering partial operators described by second order probability distributions, and more specifically Beta distributions, is introduced. An estimate to predict the second order probability of new data, obtained by averaging the second order distributions of partial operators, is derived. The problem of managing the partiality of the operators is presented. A simplistic solution based on estimating the Kolmogorov complexity of perfect completions of partial operators is given.

Nil Geisweiller

Solving Tree Problems with Category Theory

Abstract

Artificial Intelligence (AI) has long pursued models, theories, and techniques to imbue machines with human-like general intelligence. Yet even the currently predominant data-driven approasches in AI seem to be lacking humans’ unique ability to solve wide ranges of problems. This situation begs the question of the existence of principles that underlie general problem-solving capabilities. We approach this question through the mathematical formulation of analogies across different problems and solutions. We focus in particular on problems that could be represented as tree-like structures. Most importantly, we adopt a category-theoretic approach in formalising tree problems as categories, and in proving the existence of equivalences across apparently unrelated problem domains. We prove the existence of a functor between the category of tree problems and the category of solutions. We also provide a weaker version of the functor by quantifying equivalences of problem categories using a metric on tree problems.

Rafik Hadfi

Goal-Directed Procedure Learning

Abstract

A novel method of Goal-directed Procedure Learning is presented that overcomes some of the drawbacks of the traditional approaches to planning and reinforcement learning. The necessary principles for acquiring goal-dependent behaviors, and the motivations behind this approach are explained. A concrete implementation exists in a Non-Axiomatic Reasoning System, OpenNARS, although we believe the findings may be generally applicable to other AGI systems.

Patrick Hammer, Tony Lofthouse

Can Machines Design? An Artificial General Intelligence Approach

Abstract

Can machines design? Can they come up with creative solutions to problems and build tools and artifacts across a wide range of domains? Recent advances in the field of computational creativity and formal Artificial General Intelligence (AGI) provide frameworks towards machines with the general ability to design. In this paper we propose to integrate a formal computational creativity framework into the Gödel machine framework. We call the resulting framework design Gödel machine. Such a machine could solve a variety of design problems by generating novel concepts. In addition, it could change the way these concepts are generated by modifying itself. The design Gödel machine is able to improve its initial design program, once it has proven that a modification would increase its return on the utility function. Finally, we sketch out a specific version of the design Gödel machine which specifically aims at the design of complex software and hardware systems. Future work aims at the development of a more formal version of the design Gödel machine and a proof of concept implementation.

Andreas M. Hein, Hélène Condat

Resource-Constrained Social Evidence Based Cognitive Model for Empathy-Driven Artificial Intelligence

Abstract

Working model of social aspects of human and non-human intelligence is required for social embodiment of artificial general intelligence systems to explain, predict and manage behavioral patterns in multi-agent communities. For this purpose, we propose implementation of resource-constrained social evidence based model and discuss possible implications of its application.

Anton Kolonin

Unsupervised Language Learning in OpenCog

Abstract

We discuss technology capable to learn language without supervision. While the entire goal may be too ambitious and not achievable to full extent, we explore how far we can advance grammar learning. We present the current approach employed in the open source OpenCog Artificial Intelligence Platform, describe the cognitive pipeline being constructed and present some intermediate results.

Alex Glushchenko, Andres Suarez, Anton Kolonin, Ben Goertzel, Claudia Castillo, Man Hin Leung, Oleg Baskov

Functionalist Emotion Model in NARS

Abstract

Emotions play a crucial role in different cognitive functions, such as action selection and decision-making processes. This paper describes a new appraisal model for the emotion mechanism of NARS, an AGI system. Different from the previous appraisal model where emotions are triggered by the specific context, the new appraisal evaluates the relations between the system and its goals, based on a new set of criteria, including desirability, belief, and anticipation. Our work focuses on the functions of emotions and how emotional reactions could help NARS to improve its various cognitive capacities.

Xiang Li, Patrick Hammer, Pei Wang, Hongling Xie

Towards a Sociological Conception of Artificial Intelligence

Abstract

Social sciences have been always formed and influenced by the development of society, adjusting the conceptual, methodological, and theoretical frameworks to emerging social phenomena. In recent years, with the leap in the advancement of Artificial Intelligence (AI) and the proliferation of its everyday applications, “non-human intelligent actors” are increasingly becoming part of the society. This is manifested in the evolving realms of smart home systems, autonomous vehicles, chatbots, intelligent public displays, etc. In this paper, we present a prospective research project that takes one of the pioneering steps towards establishing a “distinctively sociological” conception of AI. Its first objective is to extract the existing conceptions of AI as perceived by its technological developers and (possibly differently) by its users. In the second part, capitalizing on a set of interviews with experts from social science domains, we will explore the new imaginable conceptions of AI that do not originate from its technological possibilities but rather from societal necessities. The current formal ways of defining AI are grounded in the technological possibilities, namely machine learning methods and neural network models. But what exactly is AI as a social phenomenon, which may act on its own, can be blamed responsible for ethically problematic behavior, or even endanger people’s employment? We argue that such conceptual investigation is a crucial step for further empirical studies of phenomena related to AI’s position in current societies, but also will open up ways for critiques of new technological advancements with social consequences in mind from the outset.

Jakub Mlynář, Hamed S. Alavi, Himanshu Verma, Lorenzo Cantoni

Efficient Concept Formation in Large State Spaces

Abstract

General autonomous agents must be able to operate in previously unseen worlds with large state spaces. To operate successfully in such worlds, the agents must maintain their own models of the environment, based on concept sets that are several orders of magnitude smaller. For adaptive agents, those concept sets cannot be fixed, but must adapt continuously to new situations. This, in turn, requires mechanisms for forming and preserving those concepts that are critical to successful decision-making, while removing others. In this paper we compare four general algorithms for learning and decision-making: (i) standard Q-learning, (ii) deep Q-learning, (iii) single-agent local Q-learning, and (iv) single-agent local Q-learning with improved concept formation rules. In an experiment with a state space larger than \(2^{32}\), it was found that a single-agent local Q-learning agent with improved concept formation rules performed substantially better than a similar agent with less sophisticated concept formation rules and slightly better than a deep Q-learning agent.

Fredrik Mäkeläinen, Hampus Torén, Claes Strannegård

DSO Cognitive Architecture: Implementation and Validation of the Global Workspace Enhancement

Abstract

An enhanced DSO Cognitive Architecture design was recently introduced to augment its cognitive functions by incorporating the Global Workspace Theory. A computational implementation of this new design is described in detail in this paper. The implementation is built as a distributed system with parallel pipelines of specialised processes, executing asynchronously. Competition initiated by these processes, and facilitated by the attention mechanism and global broadcast mechanism, leads to pipelines being dynamically created and allows disconnected pipelines to influence the processing of others. To validate the implementation, it was applied to a traffic control problem and experimental results showed increase in performance gain using the enhanced cognitive architecture.

Khin Hua Ng, Zhiyuan Du, Gee Wah Ng

The Foundations of Deep Learning with a Path Towards General Intelligence

Abstract

Like any field of empirical science, AI may be approached axiomatically. We formulate requirements for a general-purpose, human-level AI system in terms of postulates. We review the methodology of deep learning, examining the explicit and tacit assumptions in deep learning research. Deep Learning methodology seeks to overcome limitations in traditional machine learning research as it combines facets of model richness, generality, and practical applicability. The methodology so far has produced outstanding results due to a productive synergy of function approximation, under plausible assumptions of irreducibility and the efficiency of back-propagation family of algorithms. We examine these winning traits of deep learning, and also observe the various known failure modes of deep learning. We conclude by giving recommendations on how to extend deep learning methodology to cover the postulates of general-purpose AI including modularity, and cognitive architecture. We also relate deep learning to advances in theoretical neuroscience research.

Eray Özkural

Zeta Distribution and Transfer Learning Problem

Abstract

We explore the relations between the zeta distribution and algorithmic information theory via a new model of the transfer learning problem. The program distribution is approximated by a zeta distribution with parameter near 1. We model the training sequence as a stochastic process. We analyze the upper temporal bound for learning a training sequence and its entropy rates, assuming an oracle for the transfer learning problem. We argue from empirical evidence that power-law models are suitable for natural processes. Four sequence models are proposed. Random typing model is like no-free lunch where transfer learning does not work. Zeta process independently samples programs from the zeta distribution. A model of common sub-programs inspired by genetics uses a database of sub-programs. An evolutionary zeta process samples mutations from Zeta distribution. The analysis of stochastic processes inspired by evolution suggest that AI may be feasible in nature, countering no-free lunch sort of arguments.

Eray Özkural

Vision System for AGI: Problems and Directions

Abstract

What frameworks and architectures are necessary to create a vision system for AGI? In this paper, we propose a formal model that states the task of perception within AGI. We show the role of discriminative and generative models in achieving efficient and general solution of this task, thus specifying the task in more detail. We discuss some existing generative and discriminative models and demonstrate their insufficiency for our purposes. Finally, we discuss some architectural dilemmas and open questions.

Alexey Potapov, Sergey Rodionov, Maxim Peterson, Oleg Scherbakov, Innokentii Zhdanov, Nikolai Skorobogatko

Semantic Image Retrieval by Uniting Deep Neural Networks and Cognitive Architectures

Abstract

Image and video retrieval by their semantic content has been an important and challenging task for years, because it ultimately requires bridging the symbolic/subsymbolic gap. Recent successes in deep learning enabled detection of objects belonging to many classes greatly outperforming traditional computer vision techniques. However, deep learning solutions capable of executing retrieval queries are still not available. We propose a hybrid solution consisting of a deep neural network for object detection and a cognitive architecture for query execution. Specifically, we use YOLOv2 and OpenCog. Queries allowing the retrieval of video frames containing objects of specified classes and specified spatial arrangement are implemented.

Alexey Potapov, Innokentii Zhdanov, Oleg Scherbakov, Nikolai Skorobogatko, Hugo Latapie, Enzo Fenoglio

The Temporal Singularity: Time-Accelerated Simulated Civilizations and Their Implications

Abstract

Provided significant future progress in artificial intelligence and computing, it may ultimately be possible to create multiple Artificial General Intelligences (AGIs), and possibly entire societies living within simulated environments. In that case, it should be possible to improve the problem solving capabilities of the system by increasing the speed of the simulation. If a minimal simulation with sufficient capabilities is created, it might manage to increase its own speed by accelerating progress in science and technology, in a way similar to the Technological Singularity. This may ultimately lead to large simulated civilizations unfolding at extreme temporal speedups, achieving what from the outside would look like a Temporal Singularity. Here we discuss the feasibility of the minimal simulation and the potential advantages, dangers, and connection to the Fermi paradox of the Temporal Singularity. The medium-term importance of the topic derives from the amount of computational power required to start the process, which could be available within the next decades, making the Temporal Singularity theoretically possible before the end of the century.

Giacomo Spigler

A Computational Theory for Life-Long Learning of Semantics

Abstract

Semantic vectors are learned from data to express semantic relationships between elements of information, for the purpose of solving and informing downstream tasks. Other models exist that learn to map and classify supervised data. However, the two worlds of learning rarely interact to inform one another dynamically, whether across types of data or levels of semantics, in order to form a unified model. We explore the research problem of learning these vectors and propose a framework for learning the semantics of knowledge incrementally and online, across multiple mediums of data, via binary vectors. We discuss the aspects of this framework to spur future research on this approach and problem.

Peter Sutor Jr., Douglas Summers-Stay, Yiannis Aloimonos

Cumulative Learning with Causal-Relational Models

Abstract

In the quest for artificial general intelligence (AGI), questions remain about what kinds of representations are needed for the kind of flexibility called for by complex environments like the physical world. A capacity for continued learning of many domains has yet to be realized, and proposals for how to achieve general performance improvement through continuous cumulative learning—while seemingly a necessary feature of any AGI—remain scarce.

In this paper we describe a cumulative learning mechanism that produces causal-relational models of its environment, to predict events and achieve goals. We show how such models, coupled with an appropriate modeling process, result in knowledge whose accuracy increases over time and can run continuously throughout the lifetime of an agent. The methods have been implemented, demonstrating learning of complex tasks and situated grammatically-correct natural language by observation. Here we focus on key theoretical principles of the modeling method and explain how effective cumulative learning is achieved.

Kristinn R. Thórisson, Arthur Talbot

Transforming Kantian Aesthetic Principles into Qualitative Hermeneutics for Contemplative AGI Agents

Abstract

This paper introduces an interdisciplinary qualitative hermeneutic approach to the engineering and computer science methodological paradigm(s) for assessing a contemplative-agent’s cognitive capabilities at a level corresponding to Artificial General Intelligence (AGI). This paper has utilized cognitive intensity levels from Kantian aesthetic philosophy to qualitatively re-address Russell and Norvig’s canonical agent-categories as they scale upward towards AGI cognition levels. These Kantian levels allow the AGI-agent designer to consider the cognitive interplay between computational representations of the imagination and reason as they relate to motivationally-nuanced teleological notions of self-interest versus disinterestedness. While the AGI level is the thematic focus, lower intensity levels are also introduced in order to set the appropriate cognitive benchmarks for higher levels corresponding to truly contemplative AGI-agents. This paper first contextualizes Kant’s analytical framework before discussing the appropriately corresponding agent-categories. This paper concludes with a brief discussion of the particular methodological and hermeneutic contribution of Kant’s aesthetic philosophical framework to the AGI domain.

Jeremy O. Turner, Steve DiPaola

Towards General Evaluation of Intelligent Systems: Using Semantic Analysis to Improve Environments in the AIQ Test

Abstract

This paper conducted a semantic analysis of environment programs that are used in the Algorithmic Intelligence Quotient test to evaluate the intelligence of agents. The analysis identified several classes of programs that are non-discriminative or contain pointless code adversely affecting the testing process. Extensions of the test were implemented and verified to reduce the proportion of problematic programs thus increasing the suitability of the Algorithmic Intelligence Quotient test as a general artificial intelligence evaluation method.

Ondřej Vadinský

Perception from an AGI Perspective

Abstract

This paper argues that according to the relevant discoveries of cognitive science, in AGI systems perception should be subjective, active, and unified with other processes. This treatment of perception is fundamentally different from the mainstream approaches in computer vision and machine learning, where perception is taken to be objective, passive, and modular. The conceptual design of perception in the AGI system NARS is introduced, where the three features are realized altogether. Some preliminary testing cases are used to show the features of this novel approach.

Pei Wang, Patrick Hammer

A Phenomenologically Justifiable Simulation of Mental Modeling

Abstract

Real-world agents need to learn how to react to their environment. To achieve this, it is crucial that they have a model of this environment that is adapted during interaction and although important aspects may be hidden. This paper presents a new type of model for partially observable environments that enables an agent to represent hidden states but can still be generated and queried in realtime. Agents can use such a model to predict the outcomes of their actions and to infer action policies. These policies turn out to be better than the optimal policy in a partially observable Markov decision process as it can be inferred, for example, by Q- or Sarsa-learning. The structure and generation of these models are motivated both by phenomenological considerations from semiotics and the philosophy of mind. The performance of these models is compared to a baseline of Markov models for prediction and interaction in partially observable environments.

Mark Wernsdorfer

A Time-Critical Simulation of Language Comprehension

Abstract

Language comprehension is usually not understood as a time-critical task. Humans, however, process language on-line, in linear time, and with a single pass over a particular instance of speech or text. This calls for a genuinely cognitive algorithmic approach to simulating language comprehension. A formal conception of language is developed, as well as a model for this conception. An algorithm is presented that generates such a model on-line and from a single pass over a text. The generated model is evaluated qualitatively, by comparing its representations to linguistic segmentations (e.g. syllables, words, sentences). Results show that the model contains synonyms and homonyms as can be found in natural language. This suggests that the algorithm is able to recognize and make consistent use of context–which is crucial to understanding in general. In addition, the underlying algorithm is evaluated against a baseline approach with similar properties. This shows that the generated model is able to capture arbitrarily extended dependencies and therefore to outperform exclusively history-based approaches.

Mark Wernsdorfer

How Failure Facilitates Success

Abstract

Robotic systems that interact with real-world environments cannot capture all the underlying patterns that govern the environment’s reactions to the system’s actions. One way to deal with this uncertainty is to describe the environment probabilistically. This paper proposes another way: Failed expectations are incorporated into a deterministic model that can describe more complex dynamics than exclusively probabilistic models can. Wrong predictions from the past are used to provide a more appropriate description of the future. Unlike previous approaches, it does not suggest that transitions between hidden states can be predicted prior to the fact. Instead, effects are considered that are impossible according to the model’s current predictions. This discrepancy enables the model to self-correct in a continual coupling with the system that it describes.

Mark Wernsdorfer

Adaptive Compressed Search

Abstract

Program-search as induction and abduction is one of the key pillars of any sufficiently advanced AGI. In this paper, we present a mechanism to search for programs given a specific bias. This bias is flexible to some degree. Another novel attribute of the mechanism is the use of compression that selects simple programs over complex ones. The complexity of the program is changing all the time over the lifetime of the agent.

Robert Wünsche

Backmatter