A Roadmap Towards Machine Intelligence

By Tomas Mikolov, Armand Joulin, Marco Baroni. 2015

122 citations recorded by the source corpus.

The development of intelligent machines is one of the biggest unsolved challenges in computer science. In this paper, we propose some fundamental properties these machines should have, focusing in particular on communication and learning. We discuss a simple environment that could be used to incrementally teach a machine the basics of natural-language-based communication, as a prerequisite to more complex interaction with human users. We also present some conjectures on the sort of algorithms the machine should support in order to profitably learn from the environment.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Natural Language Processing, Reinforcement Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

In Natural Language Processing

In Reinforcement Learning

Citations in the corpus

Grounded Language Learning in a Simulated 3D World (2017, background): “Other recently-proposed frameworks for interactive language learning involve unimodal (text-only) settings (Narasimhan et al., 2015; Mikolov et al., 2015).”
ShapeWorld - A new test methodology for multimodal language understanding (2017, background): “The long-term research proposal of Mikolov et al. (2015) also simulates a world where an agent learns to solve tasks by communication with a teacher module.”
Dialog-based Language Learning (2016, background): “Recently, [14] also proposed a reward-based learning framework for learning how to learn.”
What can the brain teach us about building artificial intelligence? (2016, background): “Language also facilitates more powerful learning-to-learn and compositionality (Mikolov et al., 2016), allowing people to learn more quickly and flexibly by representing new concepts and thoughts in relation to existing concepts (Lupyan & Bergen, 2016; Lupyan & Clark, 2015).”
ParlAI: A Dialog Research Software Platform (2017, background): “CommAI7 is a framework that uses textual communication for the goal of developing artificial general intelligence through incremental tasks that test increasingly more complex skills, as described in (Mikolov et al., 2015).”
Gradient Episodic Memory for Continual Learning (2017, background): “The CommAI project [Mikolov et al., 2015, Baroni et al., 2017] shares our same motivations, but focuses on highly structured task descriptors, such as strings of text.”
Multi-Agent Cooperation and the Emergence of (Natural) Language (2016, background): “A third branch of research focuses on “Wizard-of-Oz” environments, where agents learn to play games by interacting with a complex scripted environment (Mikolov et al., 2015).”
Active Long Term Memory Networks (2016, background): “The recent interest in bridging human and machine computations [5, 6, 7, 8] obliges to consider a learning framework that is continual, sequential in nature and potentially lifelong [9, 10].”
Learning to Speak and Act in a Fantasy Text Adventure Game (2019, background): “Several position papers have proposed virtual embodiment as a strategy for language research (Brooks, 1991; Kiela et al., 2016; Gauthier and Mordatch, 2016; Mikolov et al., 2016; Lake et al., 2017).”
Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent (2017, neutral): “Grounding language in an interactive environment is an active area of research, however a number of recent works employ synthetic, templated language only (Sukhbaatar et al., 2015; Yu et al., 2017; Bordes et al., 2010; Hermann et al., 2017; Mikolov et al., 2015; Chaplot et al., 2017).”
A Paradigm for Situated and Goal-Driven Language Learning (2016, background): “Consider the following text-only interaction in the style of Mikolov et al. (2015) between a learning agent A and omniscient agent B , situated in some simulated physical world: The agent could have avoided this disaster if it had queried to ﬁnd out that the table was missing a leg before placing…”
Generalization without Systematicity: On the Compositional Skills of Sequence-to-Sequence Recurrent Networks (2017, methodology): “We call our data set SCAN because it is a S impliﬁed version of the C omm A I N avigation tasks (Mikolov et al., 2016).”
Formal Distributional Semantics: Introduction to the Special Issue (2016, background): “…Linguistics should work on general language understanding, that is, on theories and models that account both for language use as a whole, and for the associated conceptual apparatus (Collobert and Weston 2008; Mikolov, Joulin, and Baroni 2015; Goodman, Tenenbaum, and Gerstenberg 2015; Erk 2016).”
Why Build an Assistant in Minecraft? (2019, methodology): “[61] Tomas Mikolov, Armand Joulin, and Marco Baroni.”
Adversarial Language Games for Advanced Natural Language Intelligence (2019, background): “Such advanced language intelligence can be better achieved in interactive language games (Mikolov et al., 2016).”
Word meaning in minds and machines (2020, background): “…grounded in the ways discussed in the ﬁ rst desideratum, but new efforts have focused on more grounded forms of dialog through curriculum-driven language learning (Mikolov et al., 2016), text-based adventure games (Urbanek et al., 2019),and discussions of natural images (Shuster et al., 2020).”
From Particular to General: A Preliminary Case Study of Transfer Learning in Reading Comprehension (2016, methodology): “When we would like to extend it to a reinforcement learning setting the tasks accompanying A Roadmap towards Machine Intelligence [14] might replace bAbI tasks used in our case study.”
How to Motivate Your Dragon: Teaching Goal-Driven Agents to Speak and Act in Fantasy Worlds (2020, background): “One possible reason for this is that many datasets and tasks used for NLP are static, not supporting interaction and language grounding (Brooks, 1991; Feldman and Narayanan, 2004; Barsalou, 2008; Mikolov et al., 2016; Gauthier and Mordatch, 2016; Lake et al., 2017).”