Enhancing Stock Trading Performance with Deep Q-Learning by Addressing Noisy Data through Advanced Denoising Techniques
Bibliographic record
Abstract
This study presents a comparative analysis of different reinforcement learning configurations for the stock trading problem, with IBM as a case study. To address the challenge of noisy data, we explore the effectiveness of various denoising methods, including Wavelet Transform, Temporal Attention Network (TAN), and Fourier Transform, in improving model performance. We employ two different architectures, Multilayer Perceptron (MLP) and Long Short-Term Memory (LSTM), to calculate Q-values for each possible action, resulting in six distinct configurations. Evaluation is based on key metrics such as yearly returns, Sharpe ratios, and maximum drawdowns over a specified timeframe. We compare the performance of our models against benchmark strategies including Relative Strength Index (RSI), Moving Average Convergence Divergence (MACD), and Bollinger Bands. Our results demonstrate that DQN-based trading outperforms benchmark methods. Furthermore, configurations utilizing TAN, whether in conjunction with MLP or LSTM, consistently exhibit superior performance. These findings suggest that TAN-based denoising methods combined with DQN offer promising solutions for enhancing stock trading strategies using reinforcement learning techniques.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".