• The Transactions of the Korea Information Processing Society
• /
• v.6 no.5
• /
• pp.1303-1311
• /
• 1999

In order to learn in dynamic environments, reinforcement learning algorithms like Q-learning, TD(0)-learning, TD(λ)-learning have been proposed. however, most of them have a drawback of very slow learning because the reinforcement value is given when they reach their goal state. In this thesis, we have proposed a reinforcement learning method that can approximate fast to the goal state in maze environments. The proposed reinforcement learning method is separated into global learning and local learning, and then it executes learning. Global learning is a learning that uses the replacing eligibility trace method to search the goal state. In local learning, it propagates the goal state value that has been searched through global learning to neighboring sates, and then searches goal state in neighboring states. we can show through experiments that the reinforcement learning method proposed in this thesis can find out an optimal solution faster than other reinforcement learning methods like Q-learning, TD(o)learning and TD(λ)-learning.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.3
• /
• pp.166-175
• /
• 2023

Reinforcement learning can be applied to a wide variety of problems. However, the fundamental limitation of reinforcement learning is that it is difficult to derive an answer within a given time because the problems in the real world are too complex. Then, with the development of neural network technology, research on deep reinforcement learning that combines deep learning with reinforcement learning is receiving lots of attention. In this paper, two types of neural networks are combined with reinforcement learning and their characteristics were compared and analyzed with existing value-based reinforcement learning algorithms. Two types of neural networks are FNN and CNN, and existing reinforcement learning algorithms are SARSA and Q-learning.

• Computers and Concrete
• /
• v.19 no.3
• /
• pp.257-262
• /
• 2017

The aim of this study is to investigate the pH threshold value for the corrosion of steel reinforcement in concrete. A method was designed to attain the pH value of the pore solution on the location of the steel in concrete. Then the pH values of the pore solution on the location of steel in concrete were changed by exposing the samples to the environment (CO25%, RH 40%) to accelerate carbonation with different periods. Based on this, the pH threshold value for the corrosion of steel reinforcement had been examined by the methods of half-cell potential and electrochemical impedance spectra (EIS). The results have indicated that the pH threshold value for the initial corrosion of steel reinforcement in concrete was 11.21. However, in the carbonated concrete, agreement among whether steel corrosion was initiatory determined by the detection methods mentioned above could be found.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.21 no.2
• /
• pp.60-68
• /
• 2017

This study was made to examine the motion characteristics according to the reinforcement of the reinforcement length and stiffener reinforcement for shear reinforcement to the foundation structure reinforced with shear reinforcement steel plate. Experimental study was made after specimen was installed on the ground as the same as in the practical site. Reinforcement lengths of the steel for shear reinforcement are divided into 1,000 mm, 1,200 mm and 1,400 mm in the specimen and as for reinforcement method of the stiffener, 4 stiffeners with interval of 100mm reinforced with the same materials as the shear reinforcement were manufactured for the experiment. Considering result of the experiment, it is expressed that no effect of the stiffener reinforcement was found and regarding the reinforcement length of shear reinforcement material the crossed point of the two converted lines of the value that the shear force is expressed in the bearing power in the expanded dangerous section and the value that the shear capacity receivable by the reinforcement materials in the dangerous section is proposed as effective reinforcement length.

• Progress in Superconductivity and Cryogenics
• /
• v.3 no.1
• /
• pp.6-10
• /
• 2001

Stress/stram dependencies of the critical current $I_c$ in AgMgNi sheathed multifilamentary Bi(2212) superconducting tapes were evaluated at 77K, 0T. The external reinforcement was accomplished by soldering Ag-Mg tapes to sin91e side or both sides of the sample. With the external reinforcement. the strength of tapes increased but $I_c$, decreased The $I_c$, degradation characteristic according to the external reinforcement was improved markedly in terms of the stress although it appeared less rectal.table on the basis of the strain. Effects of external reinforcement were discussed in a viewpoint of monitoring sensitivity of cracking in superconducting filaments by considering n-value representing the transport behavior of the current. It is closely associated with the location of them relative to the voltage-monitoring region in the tape.

• Journal of Korean Association for Spatial Structures
• /
• v.21 no.2
• /
• pp.33-40
• /
• 2021

For the practical application of U-flanged Truss Hybrid beams, the flexural capacity of hybrid beams with end reinforcement details using vertical steel plates was verified. The bending test of U-flanged Truss Hybrid beams was performed using the same top chord under the compressive force, but with the thickness of the bottom plate and the amount of tensile reinforcement. The initial stiffness and maximum load of the specimen with tensile reinforcement have a higher value than that of the specimen without tension reinforcement, but the more tensile reinforcement, the greater the load decrease after the maximum load. In the case of the specimen with tensile reinforcement, because the test result value is 76% to 88% when compared with the flexural strength according to Korea Design Code, the safety of the U-flanged Truss Hybrid beam with the same details of the specimens can't ensure. Therefore, the development of new details is required to ensure that the bottom steel plate and the tensile reinforcement can undergo sufficient tensile deformation.

• Computers and Concrete
• /
• v.18 no.3
• /
• pp.337-354
• /
• 2016

To study the influence on fracture properties of reinforced concrete wedge splitting test specimens by the addition of reinforcement, and the restriction of steel bars on crack propagation, 7 groups reinforced concrete specimens of different reinforcement position and 1 group plain concrete specimens with the same size factors were designed and constructed for the tests. Based on the double-K fracture criterion and tests, fracture toughness calculation model which was suitable for reinforced concrete wedge splitting tensile specimens has been obtained. The results show that: the value of initial craking load Pini and unstable fracture load Pun decreases gradually with the distance of reinforcement away from specimens's top. Compared with plain concrete specimens, addition of steel bar can reduce the value of initial fracture toughness KIini, but significantly increase the value of the critical effective crack length ac and unstable fracture toughness KIun. For tensional concrete member, the effect of anti-cracking by reinforcement was mainly acted after cracking, the best function of preventing fracture initiation was when the steel bar was placed in the middle of the crack, and when the reinforcement was across the crack and located away from crack tip, it plays the best role in inhibiting the extension of crack.

• Proceedings of the IEEK Conference
• /
• 2001.06c
• /
• pp.13-16
• /
• 2001

In reinforcement teaming, Q-learning converges quite slowly to a good policy. Its because searching for the goal state takes very long time in a large stochastic domain. So I propose the speedup method using the Q-value initialization for model-free reinforcement learning. In the speedup method, it learns a naive model of a domain and makes boundaries around the goal state. By using these boundaries, it assigns the initial Q-values to the state-action pairs and does Q-learning with the initial Q-values. The initial Q-values guide the agent to the goal state in the early states of learning, so that Q-teaming updates Q-values efficiently. Therefore it saves exploration time to search for the goal state and has better performance than Q-learning. 1 present Speedup Q-learning algorithm to implement the speedup method. This algorithm is evaluated. in a grid-world domain and compared to Q-teaming.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1150-1156
• /
• 2019

This paper explores a model-free value-based approach for solving survival gridworld problem. Survival gridworld problem opens up a challenge involving taking risks to gain better rewards. Classic value-based approach in model-free reinforcement learning assumes minimal risk decisions. The proposed method involves a hybrid on-policy and off-policy updates to experience roll-outs using a modified Q-based update equation that introduces a parametric linear rectifier and motivational discount. The significance of this approach is it allows model-free training of agents that take into account risk factors and motivated exploration to gain better path decisions. Experimentations suggest that the proposed method achieved better exploration and path selection resulting to higher episode scores than classic off-policy and on-policy Q-based updates.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.30 no.1
• /
• pp.38-49
• /
• 1988

This paper presented as follows results of laboratory model tests with various shaped footings on soil bed reinforced with the strips on the base of behaviour of soil structure according to the loads and triaxial test results reinforced with geotextiles. Their parameters studied were the effects on the bearing capacity of a footing of the first layer of reinforcement, horizontal and vertical spacing of layers, number of layers, tensile strength of reinforcement and iclination load to the vertical 1.Depending on the strip arrangement, ultimate bearing capacity values could be more improved than urreinforced soil and the failure of soil was that the soil structure was transfered from the macrospace to microspase and its arrangement, from edge to edge to face to face. 2.The reinforcement was produced the reinforcing effects due to controlling the value of factor of one and permeable reinforcement was never a barrier of drainage condition. 3.Strength ratio was decreased as a linear shape according to increment of saturation degree of soil used even though at the lower strength ratio, the value of M-factor was rot influenced on the strength ratio but impermeable reinforcement decreased the strength of bearing capacity. 4.Ultimate bearing capacity under the plane-strain condition was appeared a little larger than triaxial or the other theoretical formulars and the circular footing more effective. 5.The maximum reinforcing effects were obtained at U I B=o.5, B / B=3 and N=3, when over that limit only acting as a anchor, and same strength of fabric appeared larger reinforcing effects compared to the thinner one. 6.As the LDR increased, more and more BCR occurred and there was appeared a block action below Z / B=O.5, but over the value, decrement of BCR was shown linear relation, and no effects above one. 7.The coefficient of the inclination was shown of minimum at the three layers of fabrics, but the value of H / B related to the ultimate load was decreased as increment of inclination degree, even though over the value of 4.5 there wasn't expected to the reinforcing effects As a consequence of the effects on load inclination, the degree of inclination of 15 per cent was decreased the bearing capacity of 70 per cent but irnproved the effects of 45 per cent through the insertion of geotextile.