Yuansong Liao and John Moody Department of Computer Science, Oregon Graduate Institute, P.O.Box 91000, Portland, OR 97291-1000 Abstract The committee approach has been proposed for reducing model uncertainty and improving generalization performance. The advantage of committees depends on (1) the performance of individual members and (2) the correlational structure of errors between members. This paper presents an input grouping technique for designing a heterogeneous committee. With this technique, all input variables are first grouped based on their mutual information. Statistically similar variables are assigned to the same group.

Yuansong Liao and John Moody Department of Computer Science, Oregon Graduate Institute, P.O.Box 91000, Portland, OR 97291-1000 Abstract The committee approach has been proposed for reducing model uncertainty and improving generalization performance. The advantage ofcommittees depends on (1) the performance of individual members and (2) the correlational structure of errors between members. This paper presents an input grouping technique for designing aheterogeneous committee. With this technique, all input variables are first grouped based on their mutual information. Statistically similarvariables are assigned to the same group.

In this paper, we propose to use recurrent reinforcement learning to directly optimize such trading system performance functions, and we compare two different reinforcement learning methods. The first, Recurrent Reinforcement Learning, uses immediate rewards to train the trading systems, while the second (Q-Learning (Watkins 1989)) approximates discounted future rewards. These methodologies can be applied to optimizing systems designed to trade a single security or to trade portfolios . In addition, we propose a novel value function for risk-adjusted return that enables learning to be done online: the differential Sharpe ratio. Trading system profits depend upon sequences of interdependent decisions, and are thus path-dependent. Optimal trading decisions when the effects of transactions costs, market impact and taxes are included require knowledge of the current system state. In Moody, Wu, Liao & Saffell (1998), we demonstrate that reinforcement learning provides a more elegant and effective means for training trading systems when transaction costs are included, than do more standard supervised approaches.

Inthis paper, we propose to use recurrent reinforcement learning to directly optimize such trading system performance functions, and we compare two different reinforcementlearning methods. The first, Recurrent Reinforcement Learning, uses immediate rewards to train the trading systems, while the second (Q-Learning (Watkins 1989)) approximates discounted future rewards. These methodologies can be applied to optimizing systems designed to trade a single security or to trade portfolios .In addition, we propose a novel value function for risk-adjusted return that enables learning to be done online: the differential Sharpe ratio. Trading system profits depend upon sequences of interdependent decisions, and are thus path-dependent. Optimal trading decisions when the effects of transactions costs, market impact and taxes are included require knowledge of the current system state. In Moody, Wu, Liao & Saffell (1998), we demonstrate that reinforcement learning provides a more elegant and effective means for training trading systems when transaction costs are included, than do more standard supervised approaches.

Smoothing regularizers for radial basis functions have been studied extensively, but no general smoothing regularizers for projective basis junctions (PBFs), such as the widely-used sigmoidal PBFs, have heretofore been proposed. We derive new classes of algebraically-simple mH'-order smoothing regularizers for networks of the form f(W, x)

High frequency foreign exchange data can be decomposed into three components: the inventory effect component, the surprise infonnation (news) component and the regular infonnation component. The presence of the inventory effect and news can make analysis of trends due to the diffusion of infonnation (regular information component) difficult. We propose a neural-net-based, independent component analysis to separate high frequency foreign exchange data into these three components. Our empirical results show that our proposed multi-effect decomposition can reveal the intrinsic price behavior.

Smoothing regularizers for radial basis functions have been studied extensively, but no general smoothing regularizers for projective basis junctions (PBFs), such as the widely-used sigmoidal PBFs, have heretofore been proposed. We derive newclasses of algebraically-simple mH'-order smoothing regularizers for networks of the form f(W, x)

High frequency foreign exchange data can be decomposed into three components: the inventory effect component, the surprise infonnation (news) component and the regular infonnation component. The presence of the inventory effect and news can make analysis of trends due to the diffusion of infonnation (regular information component) difficult. We propose a neural-net-based, independent component analysis to separate highfrequency foreign exchange data into these three components. Our empirical results show that our proposed multi-effect decomposition can reveal the intrinsic price behavior.

We derive a smoothing regularizer for recurrent network models by requiring robustness in prediction performance to perturbations of the training data. The regularizer can be viewed as a generalization of the first order Tikhonov stabilizer to dynamic models. The closed-form expression of the regularizer covers both time-lagged and simultaneous recurrent nets, with feedforward nets and onelayer linear nets as special cases. We have successfully tested this regularizer in a number of case studies and found that it performs better than standard quadratic weight decay. 1 Introd uction One technique for preventing a neural network from overfitting noisy data is to add a regularizer to the error function being minimized. Regularizers typically smooth the fit to noisy data. Well-established techniques include ridge regression, see (Hoerl & Kennard 1970), and more generally spline smoothing functions or Tikhonov stabilizers that penalize the mth-order squared derivatives of the function being fit, as in (Tikhonov & Arsenin 1977), (Eubank 1988), (Hastie & Tibshirani 1990) and (Wahba 1990). Thes(-ilethods have recently been extended to networks of radial basis functions (Girosi, Jones & Poggio 1995), and several heuristic approaches have been developed for sigmoidal neural networks, for example, quadratic weight decay (Plaut, Nowlan & Hinton 1986), weight elimination (Scalettar & Zee 1988),(Chauvin 1990),(Weigend, Rumelhart & Huberman 1990) and soft weight sharing (Nowlan & Hinton 1992).

We derive a smoothing regularizer for recurrent network models by requiring robustness in prediction performance to perturbations of the training data. The regularizer can be viewed as a generalization ofthe first order Tikhonov stabilizer to dynamic models. The closed-form expression of the regularizer covers both time-lagged and simultaneous recurrent nets, with feedforward nets and onelayer linearnets as special cases. We have successfully tested this regularizer in a number of case studies and found that it performs better than standard quadratic weight decay. 1 Introd uction One technique for preventing a neural network from overfitting noisy data is to add a regularizer to the error function being minimized. Regularizers typically smooth the fit to noisy data. Well-established techniques include ridge regression, see (Hoerl & Kennard 1970), and more generally spline smoothing functions or Tikhonov stabilizers that penalize the mth-order squared derivatives of the function being fit, as in (Tikhonov & Arsenin 1977), (Eubank 1988), (Hastie & Tibshirani 1990) and (Wahba 1990). Thes( -ilethods have recently been extended to networks of radial basis functions (Girosi, Jones & Poggio 1995), and several heuristic approaches have been developed for sigmoidal neural networks, for example, quadratic weight decay (Plaut, Nowlan & Hinton 1986), weight elimination (Scalettar & Zee 1988),(Chauvin 1990),(Weigend,Rumelhart & Huberman 1990) and soft weight sharing (Nowlan & Hinton 1992).