arXiv subjects

Taisuke Kobayashi

Publications and source records attributed to Taisuke Kobayashi.

At least 37 records · Page 2Linked to original sources

arXiv2023

Soft Actor-Critic Algorithm with Truly-satisfied Inequality Constraint

Soft actor-critic (SAC) in reinforcement learning is expected to be one of the next-generation robot control schemes. Its ability to maximize policy entropy would make a robotic controller robust to noise and perturbation, which is useful for real-world robot applications. However, the priority of maximizing the policy entropy is automatically tuned in the current implementation, the rule of which can be interpreted as one for equality constraint, binding the policy entropy into its specified lower bound. The current SAC is therefore no longer maximize the policy entropy, contrary to our expectation. To resolve this issue in SAC, this paper improves its implementation with a learnable state-dependent slack variable for appropriately handling the inequality constraint to maximize the policy entropy by reformulating it as the corresponding equality constraint. The introduced slack variable is optimized by a switching-type loss function that takes into account the dual objectives of satisfying the equality constraint and checking the lower bound. In Mujoco and Pybullet simulators, the modified SAC statistically achieved the higher robustness for adversarial attacks than before while regularizing the norm of action. A real-robot variable impedance task was demonstrated for showing the applicability of the modified SAC to real-world robot control. In particular, the modified SAC maintained adaptive behaviors for physical human-robot interaction, which had no experience at all during training. https://youtu.be/EH3xVtlVaJw

cs.LG↗

arXiv2022

Reward Bonuses with Gain Scheduling Inspired by Iterative Deepening Search

This paper introduces a novel method of adding intrinsic bonuses to task-oriented reward function in order to efficiently facilitate reinforcement learning search. While various bonuses have been designed to date, they are analogous to the depth-first and breadth-first search algorithms in graph theory. This paper, therefore, first designs two bonuses for each of them. Then, a heuristic gain scheduling is applied to the designed bonuses, inspired by the iterative deepening search, which is known to inherit the advantages of the two search algorithms. The proposed method is expected to allow agent to efficiently reach the best solution in deeper states by gradually exploring unknown states. In three locomotion tasks with dense rewards and three simple tasks with sparse rewards, it is shown that the two types of bonuses contribute to the performance improvement of the different tasks complementarily. In addition, by combining them with the proposed gain scheduling, all tasks can be accomplished with high performance.

cs.LG↗

arXiv2022

Mirror-Descent Inverse Kinematics for Box-constrained Joint Space

To control humanoid robots, the reference pose of end effector(s) is planned in task space, then mapped into the reference joints by IK. By viewing that problem as approximate quadratic programming (QP), recent QP solvers can be applied to solve it precisely, but iterative numerical IK solvers based on Jacobian are still in high demand due to their low computational cost. However, the conventional Jacobian-based IK usually clamps the obtained joints during iteration according to the constraints in practice, causing numerical instability due to non-smoothed objective function. To alleviate the clamping problem, this study explicitly considers the joint constraints, especially the box constraints in this paper, inside the new IK solver. Specifically, instead of clamping, a mirror descent (MD) method with box-constrained real joint space and no-constrained mirror space is integrated with the Jacobian-based IK, so-called MD-IK. In addition, to escape local optima nearly on the boundaries of constraints, a heuristic technique, called $ε$-clamping, is implemented as margin in software level. Finally, to increase convergence speed, the acceleration method for MD is integrated assuming continuity of solutions at each time. As a result, the accelerated MD-IK achieved more stable and enough fast tracking performance compared to the conventional IK solvers. The low computational cost of the proposed method mitigated the time delay until the solution is obtained in real-time humanoid gait control, achieving a more stable gait.

cs.RO↗

arXiv2022

Sparse Representation Learning with Modified q-VAE towards Minimal Realization of World Model

Extraction of low-dimensional latent space from high-dimensional observation data is essential to construct a real-time robot controller with a world model on the extracted latent space. However, there is no established method for tuning the dimension size of the latent space automatically, suffering from finding the necessary and sufficient dimension size, i.e. the minimal realization of the world model. In this study, we analyze and improve Tsallis-based variational autoencoder (q-VAE), and reveal that, under an appropriate configuration, it always facilitates making the latent space sparse. Even if the dimension size of the pre-specified latent space is redundant compared to the minimal realization, this sparsification collapses unnecessary dimensions, allowing for easy removal of them. We experimentally verified the benefits of the sparsification by the proposed method that it can easily find the necessary and sufficient six dimensions for a reaching task with a mobile manipulator that requires a six-dimensional state space. Moreover, by planning with such a minimal-realization world model learned in the extracted dimensions, the proposed method was able to exert a more optimal action sequence in real-time, reducing the reaching accomplishment time by around 20 %. The attached video is uploaded on youtube: https://youtu.be/-QjITrnxaRs

cs.LG↗

arXiv2022

Optimistic Reinforcement Learning by Forward Kullback-Leibler Divergence Optimization

This paper addresses a new interpretation of the traditional optimization method in reinforcement learning (RL) as optimization problems using reverse Kullback-Leibler (KL) divergence, and derives a new optimization method using forward KL divergence, instead of reverse KL divergence in the optimization problems. Although RL originally aims to maximize return indirectly through optimization of policy, the recent work by Levine has proposed a different derivation process with explicit consideration of optimality as stochastic variable. This paper follows this concept and formulates the traditional learning laws for both value function and policy as the optimization problems with reverse KL divergence including optimality. Focusing on the asymmetry of KL divergence, the new optimization problems with forward KL divergence are derived. Remarkably, such new optimization problems can be regarded as optimistic RL. That optimism is intuitively specified by a hyperparameter converted from an uncertainty parameter. In addition, it can be enhanced when it is integrated with prioritized experience replay and eligibility traces, both of which accelerate learning. The effects of this expected optimism was investigated through learning tendencies on numerical simulations using Pybullet. As a result, moderate optimism accelerated learning and yielded higher rewards. In a realistic robotic simulation, the proposed method with the moderate optimism outperformed one of the state-of-the-art RL method.

cs.LG↗

arXiv2022

Artificial Perception Meets Psychophysics, Revealing a Fundamental Law of Illusory Motion

Rotating Snakes is a visual illusion in which a stationary design is perceived to move dramatically. In the current study, the mechanism that generates perception of motion was analyzed using a combination of psychophysics experiments and deep neural network models that mimic human vision. We prepared three- and four-color illusion-like designs with a wide range of luminance and measured their strength of induced rotational motion. As a result, we discovered the fundamental law that the effect of the four-color snake rotation illusion was successfully enhanced by the combination of two perceptual motion vectors produced by the two three-color designs. In years to come, deep neural network technology will be one of the most effective tools not only for engineering applications but also for human perception research.

q-bio.NC↗

arXiv2022

Proximal Policy Optimization with Adaptive Threshold for Symmetric Relative Density Ratio

Deep reinforcement learning (DRL) is one of the promising approaches for introducing robots into complicated environments. The recent remarkable progress of DRL stands on regularization of policy, which allows the policy to improve stably and efficiently. A popular method, so-called proximal policy optimization (PPO), and its variants constrain density ratio of the latest and baseline policies when the density ratio exceeds a given threshold. This threshold can be designed relatively intuitively, and in fact its recommended value range has been suggested. However, the density ratio is asymmetric for its center, and the possible error scale from its center, which should be close to the threshold, would depend on how the baseline policy is given. In order to maximize the values of regularization of policy, this paper proposes a new PPO derived using relative Pearson (RPE) divergence, therefore so-called PPO-RPE, to design the threshold adaptively. In PPO-RPE, the relative density ratio, which can be formed with symmetry, replaces the raw density ratio. Thanks to this symmetry, its error scale from center can easily be estimated, hence, the threshold can be adapted for the estimated error scale. From three simple benchmark simulations, the importance of algorithm-dependent threshold design is revealed. By simulating additional four locomotion tasks, it is verified that the proposed method statistically contributes to task accomplishment by appropriately restricting the policy updates.

cs.LG↗

arXiv2022

Consolidated Adaptive T-soft Update for Deep Reinforcement Learning

Demand for deep reinforcement learning (DRL) is gradually increased to enable robots to perform complex tasks, while DRL is known to be unstable. As a technique to stabilize its learning, a target network that slowly and asymptotically matches a main network is widely employed to generate stable pseudo-supervised signals. Recently, T-soft update has been proposed as a noise-robust update rule for the target network and has contributed to improving the DRL performance. However, the noise robustness of T-soft update is specified by a hyperparameter, which should be tuned for each task, and is deteriorated by a simplified implementation. This study develops adaptive T-soft (AT-soft) update by utilizing the update rule in AdaTerm, which has been developed recently. In addition, the concern that the target network does not asymptotically match the main network is mitigated by a new consolidation for bringing the main network back to the target network. This so-called consolidated AT-soft (CAT-soft) update is verified through numerical simulations.

cs.LG↗

arXiv2022

L2C2: Locally Lipschitz Continuous Constraint towards Stable and Smooth Reinforcement Learning

This paper proposes a new regularization technique for reinforcement learning (RL) towards making policy and value functions smooth and stable. RL is known for the instability of the learning process and the sensitivity of the acquired policy to noise. Several methods have been proposed to resolve these problems, and in summary, the smoothness of policy and value functions learned mainly in RL contributes to these problems. However, if these functions are extremely smooth, their expressiveness would be lost, resulting in not obtaining the global optimal solution. This paper therefore considers RL under local Lipschitz continuity constraint, so-called L2C2. By designing the spatio-temporal locally compact space for L2C2 from the state transition at each time step, the moderate smoothness can be achieved without loss of expressiveness. Numerical noisy simulations verified that the proposed L2C2 outperforms the task performance while smoothing out the robot action generated from the learned policy.

cs.RO↗

arXiv2022

Motion Illusion-like Patterns Extracted from Photo and Art Images Using Predictive Deep Neural Networks

In our previous study, we successfully reproduced the illusory motion of the rotating snakes illusion using deep neural networks incorporating predictive coding theory. In the present study, we further examined the properties of the networks using a set of 1500 images, including ordinary static images of paintings and photographs and images of various types of motion illusions. Results showed that the networks clearly classified illusory images and others and reproduced illusory motions against various types of illusions similar to human perception. Notably, the networks occasionally detected anomalous motion vectors, even in ordinally static images where humans were unable to perceive any illusory motion. Additionally, illusion-like designs with repeating patterns were generated using areas where anomalous vectors were detected, and psychophysical experiments were conducted, in which illusory motion perception in the generated designs was detected. The observed inaccuracy of the networks will provide useful information for further understanding information processing associated with human vision.

q-bio.NC↗

arXiv2022

Adaptive and Multiple Time-scale Eligibility Traces for Online Deep Reinforcement Learning

Deep reinforcement learning (DRL) is one promising approach to teaching robots to perform complex tasks. Because methods that directly reuse the stored experience data cannot follow the change of the environment in robotic problems with a time-varying environment, online DRL is required. The eligibility traces method is well known as an online learning technique for improving sample efficiency in traditional reinforcement learning with linear regressors rather than DRL. The dependency between parameters of deep neural networks would destroy the eligibility traces, which is why they are not integrated with DRL. Although replacing the gradient with the most influential one rather than accumulating the gradients as the eligibility traces can alleviate this problem, the replacing operation reduces the number of reuses of previous experiences. To address these issues, this study proposes a new eligibility traces method that can be used even in DRL while maintaining high sample efficiency. When the accumulated gradients differ from those computed using the latest parameters, the proposed method takes into account the divergence between the past and latest parameters to adaptively decay the eligibility traces. Bregman divergences between outputs computed by the past and latest parameters are exploited due to the infeasible computational cost of the divergence between the past and latest parameters. In addition, a generalized method with multiple time-scale traces is designed for the first time. This design allows for the replacement of the most influential adaptively accumulated (decayed) eligibility traces.

cs.RO↗

arXiv2021

Towards Autonomous Driving of Personal Mobility with Small and Noisy Dataset using Tsallis-statistics-based Behavioral Cloning

Autonomous driving has made great progress and been introduced in practical use step by step. On the other hand, the concept of personal mobility is also getting popular, and its autonomous driving specialized for individual drivers is expected for a new step. However, it is difficult to collect a large driving dataset, which is basically required for the learning of autonomous driving, from the individual driver of the personal mobility. In addition, when the driver is not familiar with the operation of the personal mobility, the dataset will contain non-optimal data. This study therefore focuses on an autonomous driving method for the personal mobility with such a small and noisy, so-called personal, dataset. Specifically, we introduce a new loss function based on Tsallis statistics that weights gradients depending on the original loss function and allows us to exclude noisy data in the optimization phase. In addition, we improve the visualization technique to verify whether the driver and the controller have the same region of interest. From the experimental results, we found that the conventional autonomous driving failed to drive properly due to the wrong operations in the personal dataset, and the region of interest was different from that of the driver. In contrast, the proposed method learned robustly against the errors and successfully drove automatically while paying attention to the similar region to the driver. Attached video is also uploaded on youtube: https://youtu.be/KEq8-bOxYQA

cs.LG↗

arXiv2021

Impact of GPU uncertainty on the training of predictive deep neural networks

[retracted] We found out that the difference was dependent on the Chainer library, and does not replicate with another library (pytorch) which indicates that the results are probably due to a bug in Chainer, rather than being hardware-dependent. -- old abstract Deep neural networks often present uncertainties such as hardware- and software-derived noise and randomness. We studied the effects of such uncertainty on learning outcomes, with a particular focus on the function of graphics processing units (GPUs), and found that GPU-induced uncertainty increased learning accuracy of a certain deep neural network. When training a predictive deep neural network using only the CPU without the GPU, the learning error is higher than when training the same number of epochs using the GPU, suggesting that the GPU plays a different role in the learning process than just increasing the computational speed. Because this effect cannot be observed in learning by a simple autoencoder, it could be a phenomenon specific to certain types of neural networks. GPU-specific computational processing is more indeterminate than that by CPUs, and hardware-derived uncertainties, which are often considered obstacles that need to be eliminated, might, in some cases, be successfully incorporated into the training of deep neural networks. Moreover, such uncertainties might be interesting phenomena to consider in brain-related computational processing, which comprises a large mass of uncertain signals.

cs.LG↗

arXiv2021

q-VAE for Disentangled Representation Learning and Latent Dynamical Systems

A variational autoencoder (VAE) derived from Tsallis statistics called q-VAE is proposed. In the proposed method, a standard VAE is employed to statistically extract latent space hidden in sampled data, and this latent space helps make robots controllable in feasible computational time and cost. To improve the usefulness of the latent space, this paper focuses on disentangled representation learning, e.g., $β$-VAE, which is the baseline for it. Starting from a Tsallis statistics perspective, a new lower bound for the proposed q-VAE is derived to maximize the likelihood of the sampled data, which can be considered an adaptive $β$-VAE with deformed Kullback-Leibler divergence. To verify the benefits of the proposed q-VAE, a benchmark task to extract the latent space from the MNIST dataset was performed. The results demonstrate that the proposed q-VAE improved disentangled representation while maintaining the reconstruction accuracy of the data. In addition, it relaxes the independency condition between data, which is demonstrated by learning the latent dynamics of nonlinear dynamical systems. By combining disentangled representation, the proposed q-VAE achieves stable and accurate long-term state prediction from the initial state and the action sequence. The dataset for hexapod walking is available on IEEE Dataport, doi: https://dx.doi.org/10.21227/99af-jw71.

cs.LG↗

arXiv2021

Adaptive t-Momentum-based Optimization for Unknown Ratio of Outliers in Amateur Data in Imitation Learning

Behavioral cloning (BC) bears a high potential for safe and direct transfer of human skills to robots. However, demonstrations performed by human operators often contain noise or imperfect behaviors that can affect the efficiency of the imitator if left unchecked. In order to allow the imitators to effectively learn from imperfect demonstrations, we propose to employ the robust t-momentum optimization algorithm. This algorithm builds on the Student's t-distribution in order to deal with heavy-tailed data and reduce the effect of outlying observations. We extend the t-momentum algorithm to allow for an adaptive and automatic robustness and show empirically how the algorithm can be used to produce robust BC imitators against datasets with unknown heaviness. Indeed, the imitators trained with the t-momentum-based Adam optimizers displayed robustness to imperfect demonstrations on two different manipulation tasks with different robots and revealed the capability to take advantage of the additional data while reducing the adverse effect of non-optimal behaviors.

cs.LG↗

arXiv2021

Latent Representation in Human-Robot Interaction with Explicit Consideration of Periodic Dynamics

This paper presents a new data-driven framework for analyzing periodic physical human-robot interaction (pHRI) in latent state space. To elaborate human understanding and/or robot control during pHRI, the model representing pHRI is critical. Recent developments of deep learning technologies would enable us to learn such a model from a dataset collected from the actual pHRI. Our framework is developed based on variational recurrent neural network (VRNN), which can inherently handle time-series data like one pHRI generates. This paper modifies VRNN in order to include the latent dynamics from robot to human explicitly. In addition, to analyze periodic motions like walking, we integrate a new recurrent network based on reservoir computing (RC), which has random and fixed connections between numerous neurons, with VRNN. By augmenting RC into complex domain, periodic behavior can be represented as the phase rotation in complex domain without decaying the amplitude. For verification of the proposed framework, a rope-rotation/swinging experiment was analyzed. The proposed framework, trained on the dataset collected from the experiment, achieved the latent state space where the differences in periodic motions can be distinguished. Such a well-distinguished space yielded the best prediction accuracy of the human observations and the robot actions. The attached video can be seen in youtube: https://youtu.be/umn0MVcIpsY

cs.RO↗

arXiv2021

Hyperbolically-Discounted Reinforcement Learning on Reward-Punishment Framework

This paper proposes a new reinforcement learning with hyperbolic discounting. Combining a new temporal difference error with the hyperbolic discounting in recursive manner and reward-punishment framework, a new scheme to learn the optimal policy is derived. In simulations, it is found that the proposal outperforms the standard reinforcement learning, although the performance depends on the design of reward and punishment. In addition, the averages of discount factors w.r.t. reward and punishment are different from each other, like a sign effect in animal behaviors.

cs.LG↗

arXiv2021

Sample-efficient Gear-ratio Optimization for Biomechanical Energy Harvester

The biomechanical energy harvester is expected to harvest the electric energies from human motions. A tradeoff between harvesting energy and keeping the user's natural movements should be balanced via optimization techniques. In previous studies, the hardware itself has been specialized in advance for a single task like walking with constant speed on a flat. A key ingredient is Continuous Variable Transmission (CVT) to extend it applicable for multiple tasks. CVT could continuously adjust its gear ratio to balance the tradeoff for each task; however, such gear-ratio optimization problem remains open yet since its optimal solution may depend on the user, motion, and environment. Therefore, this paper focuses on a framework for data-driven optimization of a gear ratio in a CVT-equipped biomechanical energy harvester. Since the data collection requires a heavy burden on the user, we have to optimize the gear ratio for each task in the shortest possible time. To this end, our framework is designed sample-efficiently based on the fact that the user encounters multiple tasks, which are with similarities with each other. Specifically, our framework employs multi-task Bayesian optimization to reuse the optimization results of the similar tasks previously optimized by finding their similarities. Through experiments, we confirmed that, for each task, the proposed framework could achieve the optimal gear ratio of around 50~\% faster than one by random search, and that takes only around 20~minutes. Experimental results also suggested that the optimization can be accelerated by actively exploiting similarities with previously optimized tasks.

cs.RO↗