WaveCorr: Correlation-savvy Deep Reinforcement Learning for Portfolio\n Management
Bibliographic record
Abstract
The problem of portfolio management represents an important and challenging\nclass of dynamic decision making problems, where rebalancing decisions need to\nbe made over time with the consideration of many factors such as investors\npreferences, trading environments, and market conditions. In this paper, we\npresent a new portfolio policy network architecture for deep reinforcement\nlearning (DRL)that can exploit more effectively cross-asset dependency\ninformation and achieve better performance than state-of-the-art architectures.\nIn particular, we introduce a new property, referred to as \\textit{asset\npermutation invariance}, for portfolio policy networks that exploit multi-asset\ntime series data, and design the first portfolio policy network, named\nWaveCorr, that preserves this invariance property when treating asset\ncorrelation information. At the core of our design is an innovative permutation\ninvariant correlation processing layer. An extensive set of experiments are\nconducted using data from both Canadian (TSX) and American stock markets (S&P\n500), and WaveCorr consistently outperforms other architectures with an\nimpressive 3%-25% absolute improvement in terms of average annual return, and\nup to more than 200% relative improvement in average Sharpe ratio. We also\nmeasured an improvement of a factor of up to 5 in the stability of performance\nunder random choices of initial asset ordering and weights. The stability of\nthe network has been found as particularly valuable by our industrial partner.\n
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 | 0.001 | 0.002 |
| 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 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".