scholarly journals Skill-based curiosity for intrinsically motivated reinforcement learning

2019 ◽  
Vol 109 (3) ◽  
pp. 493-512 ◽  
Author(s):  
Nicolas Bougie ◽  
Ryutaro Ichise
Keyword(s):  
Reinforcement Learning ◽  
State Of The Art ◽  
Skill Learning ◽  
High Dimensional ◽  
Sequential Decision ◽  
Learning Methods ◽  
Reward Function ◽  
Intrinsic Reward ◽  
Reinforcement Learning Models ◽  
Data Efficiency

Abstract Reinforcement learning methods rely on rewards provided by the environment that are extrinsic to the agent. However, many real-world scenarios involve sparse or delayed rewards. In such cases, the agent can develop its own intrinsic reward function called curiosity to enable the agent to explore its environment in the quest of new skills. We propose a novel end-to-end curiosity mechanism for deep reinforcement learning methods, that allows an agent to gradually acquire new skills. Our method scales to high-dimensional problems, avoids the need of directly predicting the future, and, can perform in sequential decision scenarios. We formulate the curiosity as the ability of the agent to predict its own knowledge about the task. We base the prediction on the idea of skill learning to incentivize the discovery of new skills, and guide exploration towards promising solutions. To further improve data efficiency and generalization of the agent, we propose to learn a latent representation of the skills. We present a variety of sparse reward tasks in MiniGrid, MuJoCo, and Atari games. We compare the performance of an augmented agent that uses our curiosity reward to state-of-the-art learners. Experimental evaluation exhibits higher performance compared to reinforcement learning models that only learn by maximizing extrinsic rewards.

scholarly journals Logic-Based Sequential Decision-Making

2019 ◽  
Vol 33 ◽  
pp. 9995-9996
Author(s):  
Daoming Lyu ◽  
Fangkai Yang ◽  
Bo Liu ◽  
Daesub Yoon
Keyword(s):  
Decision Making ◽  
Reinforcement Learning ◽  
High Dimensional ◽  
Great Success ◽  
Sequential Decision ◽  
Sensory Inputs ◽  
Hierarchical Decision ◽  
High Level ◽  
Data Efficiency ◽  
Symbolic Planning

Deep reinforcement learning (DRL) has gained great success by learning directly from high-dimensional sensory inputs, yet is notorious for the lack of interpretability. Interpretability of the subtasks is critical in hierarchical decision-making as it increases the transparency of black-box-style DRL approach and helps the RL practitioners to understand the high-level behavior of the system better. In this paper, we introduce symbolic planning into DRL and propose a framework of Symbolic Deep Reinforcement Learning (SDRL) that can handle both high-dimensional sensory inputs and symbolic planning. The task-level interpretability is enabled by relating symbolic actions to options. This framework features a planner – controller – meta-controller architecture, which takes charge of subtask scheduling, data-driven subtask learning, and subtask evaluation, respectively. The three components cross-fertilize each other and eventually converge to an optimal symbolic plan along with the learned subtasks, bringing together the advantages of long-term planning capability with symbolic knowledge and end-to-end reinforcement learning directly from a high-dimensional sensory input. Experimental results validate the interpretability of subtasks, along with improved data efficiency compared with state-of-the-art approaches.

scholarly journals Towards High-Level Intrinsic Exploration in Reinforcement Learning

2020 ◽  
Author(s):  
Nicolas Bougie ◽  
Ryutaro Ichise
Keyword(s):  
Reinforcement Learning ◽  
Time Horizon ◽  
State Of The Art ◽  
Experimental Results ◽  
Prior Work ◽  
Extrinsic Rewards ◽  
Intrinsic Reward ◽  
Long Time ◽  
End To End ◽  
High Level

Deep reinforcement learning (DRL) methods traditionally struggle with tasks where environment rewards are sparse or delayed, which entails that exploration remains one of the key challenges of DRL. Instead of solely relying on extrinsic rewards, many state-of-the-art methods use intrinsic curiosity as exploration signal. While they hold promise of better local exploration, discovering global exploration strategies is beyond the reach of current methods. We propose a novel end-to-end intrinsic reward formulation that introduces high-level exploration in reinforcement learning. Our curiosity signal is driven by a fast reward that deals with local exploration and a slow reward that incentivizes long-time horizon exploration strategies. We formulate curiosity as the error in an agent’s ability to reconstruct the observations given their contexts. Experimental results show that this high-level exploration enables our agents to outperform prior work in several Atari games.

scholarly journals Machine Teaching for Inverse Reinforcement Learning: Algorithms and Applications

2019 ◽  
Vol 33 ◽  
pp. 7749-7758
Author(s):  
Daniel S. Brown ◽  
Scott Niekum
Keyword(s):  
Reinforcement Learning ◽  
Set Cover ◽  
Sequential Decision ◽  
Inverse Reinforcement Learning ◽  
Reward Function ◽  
Set Cover Problem ◽  
Efficient Approximation Algorithm ◽  
Minimum Number ◽  
Teaching Problem ◽  
Novel Applications

Inverse reinforcement learning (IRL) infers a reward function from demonstrations, allowing for policy improvement and generalization. However, despite much recent interest in IRL, little work has been done to understand the minimum set of demonstrations needed to teach a specific sequential decisionmaking task. We formalize the problem of finding maximally informative demonstrations for IRL as a machine teaching problem where the goal is to find the minimum number of demonstrations needed to specify the reward equivalence class of the demonstrator. We extend previous work on algorithmic teaching for sequential decision-making tasks by showing a reduction to the set cover problem which enables an efficient approximation algorithm for determining the set of maximallyinformative demonstrations. We apply our proposed machine teaching algorithm to two novel applications: providing a lower bound on the number of queries needed to learn a policy using active IRL and developing a novel IRL algorithm that can learn more efficiently from informative demonstrations than a standard IRL approach.

scholarly journals ERLP: Ensembles of Reinforcement Learning Policies (Student Abstract)

2020 ◽  
Vol 34 (10) ◽  
pp. 13905-13906
Author(s):  
Rohan Saphal ◽  
Balaraman Ravindran ◽  
Dheevatsa Mudigere ◽  
Sasikanth Avancha ◽  
Bharat Kaul
Keyword(s):  
Reinforcement Learning ◽  
State Of The Art ◽  
Multiple Models ◽  
Model Parameters ◽  
Continuous Control ◽  
Sample Complexity ◽  
Local Minima ◽  
Single Model ◽  
Learning Policies ◽  
Reinforcement Learning Models

Reinforcement learning algorithms are sensitive to hyper-parameters and require tuning and tweaking for specific environments for improving performance. Ensembles of reinforcement learning models on the other hand are known to be much more robust and stable. However, training multiple models independently on an environment suffers from high sample complexity. We present here a methodology to create multiple models from a single training instance that can be used in an ensemble through directed perturbation of the model parameters at regular intervals. This allows training a single model that converges to several local minima during the optimization process as a result of the perturbation. By saving the model parameters at each such instance, we obtain multiple policies during training that are ensembled during evaluation. We evaluate our approach on challenging discrete and continuous control tasks and also discuss various ensembling strategies. Our framework is substantially sample efficient, computationally inexpensive and is seen to outperform state of the art (SOTA) approaches

scholarly journals Deriving Subgoals Autonomously to Accelerate Learning in Sparse Reward Domains

2019 ◽  
Vol 33 ◽  
pp. 881-889 ◽  
Author(s):  
Michael Dann ◽  
Fabio Zambetta ◽  
John Thangarajah
Keyword(s):  
Reinforcement Learning ◽  
Domain Knowledge ◽  
State Of The Art ◽  
Significant Challenge ◽  
Intrinsic Reward ◽  
Art Methods ◽  
Efficient Exploration

Sparse reward games, such as the infamous Montezuma’s Revenge, pose a significant challenge for Reinforcement Learning (RL) agents. Hierarchical RL, which promotes efficient exploration via subgoals, has shown promise in these games. However, existing agents rely either on human domain knowledge or slow autonomous methods to derive suitable subgoals. In this work, we describe a new, autonomous approach for deriving subgoals from raw pixels that is more efficient than competing methods. We propose a novel intrinsic reward scheme for exploiting the derived subgoals, applying it to three Atari games with sparse rewards. Our agent’s performance is comparable to that of state-of-the-art methods, demonstrating the usefulness of the subgoals found.

scholarly journals Reward Learning for Efficient Reinforcement Learning in Extractive Document Summarisation

2019 ◽  
Author(s):  
Yang Gao ◽  
Christian M. Meyer ◽  
Mohsen Mesgar ◽  
Iryna Gurevych
Keyword(s):  
Reinforcement Learning ◽  
Learning To Rank ◽  
Poor Performance ◽  
Parameter Tuning ◽  
Test Time ◽  
Sequential Decision ◽  
Time Data ◽  
Training Time ◽  
Search Spaces ◽  
Reward Function

Document summarisation can be formulated as a sequential decision-making problem, which can be solved by Reinforcement Learning (RL) algorithms. The predominant RL paradigm for summarisation learns a cross-input policy, which requires considerable time, data and parameter tuning due to the huge search spaces and the delayed rewards. Learning input-specific RL policies is a more efficient alternative, but so far depends on handcrafted rewards, which are difficult to design and yield poor performance. We propose RELIS, a novel RL paradigm that learns a reward function with Learning-to-Rank (L2R) algorithms at training time and uses this reward function to train an input-specific RL policy at test time. We prove that RELIS guarantees to generate near-optimal summaries with appropriate L2R and RL algorithms. Empirically, we evaluate our approach on extractive multi-document summarisation. We show that RELIS reduces the training time by two orders of magnitude compared to the state-of-the-art models while performing on par with them.

scholarly journals LTL and Beyond: Formal Languages for Reward Function Specification in Reinforcement Learning

2019 ◽  
Author(s):  
Alberto Camacho ◽  
Rodrigo Toro Icarte ◽  
Toryn Q. Klassen ◽  
Richard Valenzano ◽  
Sheila A. McIlraith
Keyword(s):  
Reinforcement Learning ◽  
Normal Form ◽  
State Of The Art ◽  
Formal Languages ◽  
Function Structure ◽  
Q Learning ◽  
Reward Function ◽  
Form Representation ◽  
Reward Shaping ◽  
Reward Functions

In Reinforcement Learning (RL), an agent is guided by the rewards it receives from the reward function. Unfortunately, it may take many interactions with the environment to learn from sparse rewards, and it can be challenging to specify reward functions that reflect complex reward-worthy behavior. We propose using reward machines (RMs), which are automata-based representations that expose reward function structure, as a normal form representation for reward functions. We show how specifications of reward in various formal languages, including LTL and other regular languages, can be automatically translated into RMs, easing the burden of complex reward function specification. We then show how the exposed structure of the reward function can be exploited by tailored q-learning algorithms and automated reward shaping techniques in order to improve the sample efficiency of reinforcement learning methods. Experiments show that these RM-tailored techniques significantly outperform state-of-the-art (deep) RL algorithms, solving problems that otherwise cannot reasonably be solved by existing approaches.

scholarly journals Read, Watch, and Move: Reinforcement Learning for Temporally Grounding Natural Language Descriptions in Videos

2019 ◽  
Vol 33 ◽  
pp. 8393-8400 ◽  
Author(s):  
Dongliang He ◽  
Xiang Zhao ◽  
Jizhou Huang ◽  
Fu Li ◽  
Xiao Liu ◽  
...  
Keyword(s):  
Decision Making ◽  
Reinforcement Learning ◽  
Natural Language ◽  
State Of The Art ◽  
Sliding Window ◽  
Sequential Decision Making ◽  
Sequential Decision ◽  
Boundary Information ◽  
Performance Gains ◽  
Steady Performance

The task of video grounding, which temporally localizes a natural language description in a video, plays an important role in understanding videos. Existing studies have adopted strategies of sliding window over the entire video or exhaustively ranking all possible clip-sentence pairs in a presegmented video, which inevitably suffer from exhaustively enumerated candidates. To alleviate this problem, we formulate this task as a problem of sequential decision making by learning an agent which regulates the temporal grounding boundaries progressively based on its policy. Specifically, we propose a reinforcement learning based framework improved by multi-task learning and it shows steady performance gains by considering additional supervised boundary information during training. Our proposed framework achieves state-of-the-art performance on ActivityNet’18 DenseCaption dataset (Krishna et al. 2017) and Charades-STA dataset (Sigurdsson et al. 2016; Gao et al. 2017) while observing only 10 or less clips per video.

scholarly journals Deep intrinsically motivated continuous actor-critic for efficient robotic visuomotor skill learning

2019 ◽  
Vol 10 (1) ◽  
pp. 14-29 ◽  
Author(s):  
Muhammad Burhan Hafez ◽  
Cornelius Weber ◽  
Matthias Kerzel ◽  
Stefan Wermter
Keyword(s):  
Motor Skills ◽  
Humanoid Robot ◽  
State Of The Art ◽  
Visual Input ◽  
Skill Learning ◽  
Continuous Control ◽  
Extrinsic Reward ◽  
Neural Architecture ◽  
Intrinsic Reward ◽  
Convolutional Autoencoder

Abstract In this paper, we present a new intrinsically motivated actor-critic algorithm for learning continuous motor skills directly from raw visual input. Our neural architecture is composed of a critic and an actor network. Both networks receive the hidden representation of a deep convolutional autoencoder which is trained to reconstruct the visual input, while the centre-most hidden representation is also optimized to estimate the state value. Separately, an ensemble of predictive world models generates, based on its learning progress, an intrinsic reward signal which is combined with the extrinsic reward to guide the exploration of the actor-critic learner. Our approach is more data-efficient and inherently more stable than the existing actor-critic methods for continuous control from pixel data. We evaluate our algorithm for the task of learning robotic reaching and grasping skills on a realistic physics simulator and on a humanoid robot. The results show that the control policies learned with our approach can achieve better performance than the compared state-of-the-art and baseline algorithms in both dense-reward and challenging sparse-reward settings.

On the use of the policy gradient and Hessian in inverse reinforcement learning

2020 ◽  
Vol 14 (1) ◽  
pp. 117-150
Author(s):  
Alberto Maria Metelli ◽  
Matteo Pirotta ◽  
Marcello Restelli
Keyword(s):  
Reinforcement Learning ◽  
Sequential Decision ◽  
Inverse Reinforcement Learning ◽  
Reward Function ◽  
Model Free ◽  
Learning Speed ◽  
Policy Gradient ◽  
Continuous Domains ◽  
Learning Policies ◽  
Finite Domains

Reinforcement Learning (RL) is an effective approach to solve sequential decision making problems when the environment is equipped with a reward function to evaluate the agent’s actions. However, there are several domains in which a reward function is not available and difficult to estimate. When samples of expert agents are available, Inverse Reinforcement Learning (IRL) allows recovering a reward function that explains the demonstrated behavior. Most of the classic IRL methods, in addition to expert’s demonstrations, require sampling the environment to evaluate each reward function, that, in turn, is built starting from a set of engineered features. This paper is about a novel model-free IRL approach that does not require to specify a function space where to search for the expert’s reward function. Leveraging on the fact that the policy gradient needs to be zero for an optimal policy, the algorithm generates an approximation space for the reward function, in which a reward is singled out employing a second-order criterion. After introducing our approach for finite domains, we extend it to continuous ones. The empirical results, on both finite and continuous domains, show that the reward function recovered by our algorithm allows learning policies that outperform those obtained with the true reward function, in terms of learning speed.

Sign in / Sign up

Export Citation Format

Share Document