<s>
Google	B-Application
JAX	I-Application
is	O
a	O
machine	O
learning	O
framework	O
for	O
transforming	O
numerical	O
functions	O
.	O
</s>
<s>
It	O
is	O
described	O
as	O
bringing	O
together	O
a	O
modified	O
version	O
of	O
(	O
automatic	O
obtaining	O
of	O
the	O
gradient	O
function	O
through	O
differentiation	O
of	O
a	O
function	O
)	O
and	O
TensorFlow	B-Language
's	O
(	O
Accelerated	O
Linear	O
Algebra	O
)	O
.	O
</s>
<s>
It	O
is	O
designed	O
to	O
follow	O
the	O
structure	O
and	O
workflow	O
of	O
NumPy	B-Application
as	O
closely	O
as	O
possible	O
and	O
works	O
with	O
various	O
existing	O
frameworks	O
such	O
as	O
TensorFlow	B-Language
and	O
PyTorch	B-Algorithm
.	O
</s>
<s>
The	O
below	O
code	O
demonstrates	O
the	O
grad	O
function	O
's	O
automatic	B-Algorithm
differentiation	I-Algorithm
.	O
</s>
<s>
Several	O
python	B-Language
libraries	O
use	O
Jax	O
as	O
a	O
backend	O
,	O
including	O
:	O
</s>
<s>
Flax	O
,	O
a	O
high	O
level	O
neural	B-Architecture
network	I-Architecture
library	O
initially	O
developed	O
by	O
Google	B-Application
Brain	I-Application
.	O
</s>
<s>
Haiku	O
,	O
an	O
object-oriented	B-Language
library	O
for	O
neural	B-Architecture
networks	I-Architecture
developed	O
by	O
DeepMind	B-Application
.	O
</s>
<s>
Equinox	O
,	O
a	O
library	O
that	O
revolves	O
around	O
the	O
idea	O
of	O
representing	O
parameterised	O
functions	O
(	O
including	O
neural	B-Architecture
networks	I-Architecture
)	O
as	O
PyTrees	O
.	O
</s>
<s>
Optax	O
,	O
a	O
library	O
for	O
gradient	O
processing	O
and	O
optimisation	O
developed	O
by	O
DeepMind	B-Application
.	O
</s>
<s>
RLax	O
,	O
a	O
library	O
for	O
developing	O
reinforcement	O
learning	O
agents	O
developed	O
by	O
DeepMind	B-Application
.	O
</s>
