diff --git a/fancy_rl/norm.py b/fancy_rl/norm.py
new file mode 100644
index 0000000..f40d319
--- /dev/null
+++ b/fancy_rl/norm.py
@@ -0,0 +1,27 @@
+import torch as th
+from torch.distributions.multivariate_normal import _batch_mahalanobis
+
+
+def mahalanobis_alt(u, v, std):
+    delta = u - v
+    return th.triangular_solve(delta, std, upper=False)[0].pow(2).sum([-2, -1])
+
+
+def mahalanobis(u, v, chol):
+    delta = u - v
+    return _batch_mahalanobis(chol, delta)
+
+
+def frob_sq(diff, is_spd=False):
+    # If diff is spd, we can use a (probably) more performant algorithm
+    if is_spd:
+        return _frob_sq_spd(diff)
+    return th.norm(diff, p='fro', dim=tuple(range(1, diff.dim()))).pow(2)
+
+
+def _frob_sq_spd(diff):
+    return _batch_trace(diff @ diff)
+
+
+def _batch_trace(x):
+    return th.diagonal(x, dim1=-2, dim2=-1).sum(-1)