ترغب بنشر مسار تعليمي؟ اضغط هنا

Adaptive Reinforcement Learning through Evolving Self-Modifying Neural Networks

190   0   0.0 ( 0 )
 نشر من قبل Samuel Schmidgall
 تاريخ النشر 2020
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English
 تأليف Samuel Schmidgall




اسأل ChatGPT حول البحث

The adaptive learning capabilities seen in biological neural networks are largely a product of the self-modifying behavior emerging from online plastic changes in synaptic connectivity. Current methods in Reinforcement Learning (RL) only adjust to new interactions after reflection over a specified time interval, preventing the emergence of online adaptivity. Recent work addressing this by endowing artificial neural networks with neuromodulated plasticity have been shown to improve performance on simple RL tasks trained using backpropagation, but have yet to scale up to larger problems. Here we study the problem of meta-learning in a challenging quadruped domain, where each leg of the quadruped has a chance of becoming unusable, requiring the agent to adapt by continuing locomotion with the remaining limbs. Results demonstrate that agents evolved using self-modifying plastic networks are more capable of adapting to complex meta-learning learning tasks, even outperforming the same network updated using gradient-based algorithms while taking less time to train.



قيم البحث

اقرأ أيضاً

82 - Samuel Schmidgall 2021
The search for neural architecture is producing many of the most exciting results in artificial intelligence. It has increasingly become apparent that task-specific neural architecture plays a crucial role for effectively solving problems. This paper presents a simple method for learning neural architecture through random mutation. This method demonstrates 1) neural architecture may be learned during the agents lifetime, 2) neural architecture may be constructed over a single lifetime without any initial connections or neurons, and 3) architectural modifications enable rapid adaptation to dynamic and novel task scenarios. Starting without any neurons or connections, this method constructs a neural architecture capable of high-performance on several tasks. The lifelong learning capabilities of this method are demonstrated in an environment without episodic resets, even learning with constantly changing morphology, limb disablement, and changing task goals all without losing locomotion capabilities.
The impressive lifelong learning in animal brains is primarily enabled by plastic changes in synaptic connectivity. Importantly, these changes are not passive, but are actively controlled by neuromodulation, which is itself under the control of the b rain. The resulting self-modifying abilities of the brain play an important role in learning and adaptation, and are a major basis for biological reinforcement learning. Here we show for the first time that artificial neural networks with such neuromodulated plasticity can be trained with gradient descent. Extending previous work on differentiable Hebbian plasticity, we propose a differentiable formulation for the neuromodulation of plasticity. We show that neuromodulated plasticity improves the performance of neural networks on both reinforcement learning and supervised learning tasks. In one task, neuromodulated plastic LSTMs with millions of parameters outperform standard LSTMs on a benchmark language modeling task (controlling for the number of parameters). We conclude that differentiable neuromodulation of plasticity offers a powerful new framework for training neural networks.
Unsupervised anomaly discovery in stream data is a research topic with many practical applications. However, in many cases, it is not easy to collect enough training data with labeled anomalies for supervised learning of an anomaly detector in order to deploy it later for identification of real anomalies in streaming data. It is thus important to design anomalies detectors that can correctly detect anomalies without access to labeled training data. Our idea is to adapt the Online evolving Spiking Neural Network (OeSNN) classifier to the anomaly detection task. As a result, we offer an Online evolving Spiking Neural Network for Unsupervised Anomaly Detection algorithm (OeSNN-UAD), which, unlike OeSNN, works in an unsupervised way and does not separate output neurons into disjoint decision classes. OeSNN-UAD uses our proposed new two-step anomaly detection method. Also, we derive new theoretical properties of neuronal model and input layer encoding of OeSNN, which enable more effective and efficient detection of anomalies in our OeSNN-UAD approach. The proposed OeSNN-UAD detector was experimentally compared with state-of-the-art unsupervised and semi-supervised detectors of anomalies in stream data from the Numenta Anomaly Benchmark and Yahoo Anomaly Datasets repositories. Our approach outperforms the other solutions provided in the literature in the case of data streams from the Numenta Anomaly Benchmark repository. Also, in the case of real data files of the Yahoo Anomaly Benchmark repository, OeSNN-UAD outperforms other selected algorithms, whereas in the case of Yahoo Anomaly Benchmark synthetic data files, it provides competitive results to the results recently reported in the literature.
A Pseudo-Random Number Generator (PRNG) is any algorithm generating a sequence of numbers approximating properties of random numbers. These numbers are widely employed in mid-level cryptography and in software applications. Test suites are used to ev aluate PRNGs quality by checking statistical properties of the generated sequences. These sequences are commonly represented bit by bit. This paper proposes a Reinforcement Learning (RL) approach to the task of generating PRNGs from scratch by learning a policy to solve a partially observable Markov Decision Process (MDP), where the full state is the period of the generated sequence and the observation at each time step is the last sequence of bits appended to such state. We use a Long-Short Term Memory (LSTM) architecture to model the temporal relationship between observations at different time steps, by tasking the LSTM memory with the extraction of significant features of the hidden portion of the MDPs states. We show that modeling a PRNG with a partially observable MDP and a LSTM architecture largely improves the results of the fully observable feedforward RL approach introduced in previous work.
We present a neural architecture search algorithm to construct compact reinforcement learning (RL) policies, by combining ENAS and ES in a highly scalable and intuitive way. By defining the combinatorial search space of NAS to be the set of different edge-partitionings (colorings) into same-weight classes, we represent compact architectures via efficient learned edge-partitionings. For several RL tasks, we manage to learn colorings translating to effective policies parameterized by as few as $17$ weight parameters, providing >90% compression over vanilla policies and 6x compression over state-of-the-art compact policies based on Toeplitz matrices, while still maintaining good reward. We believe that our work is one of the first attempts to propose a rigorous approach to training structured neural network architectures for RL problems that are of interest especially in mobile robotics with limited storage and computational resources.

الأسئلة المقترحة

التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا